Rotating electric machine control system

Abstract

A rotating electric machine control system (10) has a basic current command setting section (1) that sets a basic current command (Idq*) based on a target torque (T*) of a rotating electric machine, a correction current command setting section (6) that sets a correction current command (Idqh*) of a frequency corresponding to a torque fluctuation, and a current control section (2) that performs current feedback control to calculate a command value of a voltage applied to the rotating electric machine (80), i.e., a voltage command (Vdq*). The current control section (2) has a first current control section (20) that takes components of the basic current command (Idq*) as control targets and a second current control section (21) that takes components of the correction current command (Idqh*) as control targets. The gain of the second current control section (21) is set to be lower than the gain of the first current control section (20).

Description

Technical Field

[0001] This invention relates to a rotary motor control system that uses a permanent magnet type rotary motor as the controlled object and performs current feedback control in an orthogonal vector coordinate system of the dq axes. Background Technology

[0002] The paper "Torque Ripple Reduction in Sensorless PMSM Drives" by Antti Piippo and Jorma Luomi demonstrates a technique for suppressing torque ripple in rotating electrical machines. This involves superimposing a corrective current command, representing a higher harmonic frequency corresponding to the torque ripple, onto a current command. The corrective current command is a current command that outputs a corrective torque in the opposite phase to the torque ripple, thereby reducing the torque ripple. In the dq-axis orthogonal vector coordinate system, the current command is direct current (DC), while the corrective current command is alternating current (AC). Therefore, in a controller that controls a conventional current command without the superimposed corrective current command, it may not be possible to properly control a current command with the superimposed corrective current command. In the aforementioned paper, responsiveness is improved by adding a controller corresponding to the AC current in parallel with a conventional controller.

[0003] Existing technical documents

[0004] Non-patent literature

[0005] Non-patent document 1: Antti Piippo, Jorma Luomi, "Torque Ripple Reduction in Sensorless PMSM Drives.", The 32nd Annual Conference of the IEEE Industrial Electronic Society (IECON'06), pp.920-925 Summary of the Invention

[0006] The technical problem that the invention aims to solve

[0007] The techniques described in the aforementioned paper work very effectively when the torque of the rotating electric machine is in a constant, steady state. That is, the current command, superimposed with a correction current command, is properly controlled, effectively suppressing torque fluctuations. However, when the rotating electric machine is, for example, a driving force source for a vehicle, the additional controller may sometimes react to the AC components of the rotating electric machine's torque during rapid (stepped) or continuous (scanning) changes. As a result, this may induce overshoot or vibration in the rotating electric machine's torque, a phenomenon known as control disturbance.

[0008] Given the above background, it is desirable to provide a technology that can appropriately reduce torque fluctuations without causing control disturbances even in rotating electric machines with torque variations.

[0009] means for solving problems

[0010] In view of the above-described rotating electric machine control system, a permanent magnet type rotating electric machine driven by N-phase AC (N being any natural number) is used as the controlled object. Current feedback control is performed in a dq-axis orthogonal vector coordinate system along the d-axis of the direction of the excitation flux generated by the permanent magnet and the q-axis orthogonal to the d-axis. The rotating electric machine control system includes: a basic current command setting unit, which sets a basic current command, i.e., a current command, as a command value for the current flowing in the rotating electric machine based on the target torque of the rotating electric machine; and a correction current command setting unit, which sets a correction current command, which is designed to reduce torque fluctuations in the rotating electric machine in conjunction with the basic current command. The system includes a superimposed current command, which is an alternating current at a frequency corresponding to the torque fluctuation of the object being reduced; and a current control unit that performs current feedback control based on the deviation between the object current command of the controlled object obtained by adding the basic current command and the correction current command and the current flowing through the rotating motor, i.e., the actual current, and calculates the command value of the voltage applied to the rotating motor, i.e., the voltage command. The current control unit has a first current control unit that uses the component of the basic current command as the controlled object, and a second current control unit that uses the component of the correction current command as the controlled object, wherein the gain of the second current control unit is set to be lower than the gain of the first current control unit.

[0011] The component of the basic current command controlled by the first current control unit is a direct current in the orthogonal vector coordinate system of the dq axes, while the component of the correction current command controlled by the second current control unit is an alternating current in the orthogonal vector coordinate system of the dq axes. By having a first current control unit and a second current control unit with different control objects, the basic current command superimposed with the correction current command can be used as the control object for appropriate current control. On the other hand, when the target torque changes drastically, the basic current command also changes drastically in response. Since this point of change contains high-frequency components, the second current control unit also performs current control based on the basic current command, which may result in overshoot or vibration in the output voltage command. In other words, control interference may occur. However, according to this structure, since the gain of the second current is set lower than that of the first current control unit, such overshoot is suppressed, and vibration is also suppressed. That is, according to this structure, even in a rotating motor with torque fluctuations, control interference is not generated, thereby appropriately reducing torque ripple.

[0012] Other features and advantages of the rotary electric motor control system will become clear from the following description of the embodiments with reference to the accompanying drawings. Attached Figure Description

[0013] Figure 1 This is a block diagram schematically showing the structure of a rotary electric motor drive device.

[0014] Figure 2 This is a block diagram schematically showing the structure of the rotary motor control device, which is the core of the rotary motor drive.

[0015] Figure 3 This is a schematic control block diagram showing the area surrounding the current control section in a rotating electric machine control device.

[0016] Figure 4 This is a waveform diagram showing an example of the actual torque of a rotating electric motor.

[0017] Figure 5 This is a waveform diagram illustrating an example of torque reduction caused by the oscillation component of higher harmonics and the oscillation of opposite phase.

[0018] Figure 6 This is a graph showing the relationship between torque and current commands in a Cartesian coordinate system along the dq axis.

[0019] Figure 7 It is a graph showing the relationship between superimposed current amplitude and torque.

[0020] Figure 8 It is a graph showing the relationship between the superimposed current phase and torque.

[0021] Figure 9 This is a diagram showing the operating range of a rotary electric motor based on torque and speed.

[0022] Figure 10 This is a graph showing the relationship between the reduction effect of torque ripple and the phase difference.

[0023] Figure 11 This is a characteristic diagram showing the relationship between frequency and gain in the current control section.

[0024] Figure 12 This is a waveform diagram illustrating an example of the relationship between torque command and output torque.

[0025] Figure 13 This is a characteristic diagram showing the relationship between frequency and gain in the current control section.

[0026] Figure 14 This is a characteristic diagram showing the relationship between frequency and gain in the current control section.

[0027] Figure 15This is a waveform diagram showing an example of the dq-axis current without superimposed higher harmonics.

[0028] Figure 16 This is a waveform diagram showing an example of the dq-axis current under the condition of superimposed higher harmonics. Detailed Implementation

[0029] The following describes an implementation of the rotary electric motor control system based on the accompanying drawings. The rotary electric motor control system, for example, uses a rotary electric motor that serves as the driving force source for a vehicle as the controlled object for current feedback control. Figure 1 The block diagram schematically illustrates the system structure of the rotary motor drive unit 100, including the rotary motor control unit 10 (MG-CTRL). Additionally, Figure 2 The block diagram schematically illustrates the system structure of the rotary electric motor control device 10, which is the core of the rotary electric motor drive device 100. Additionally, Figure 3 The control block diagram shows a schematic structure around the current control unit 2 in the rotary electric machine control device 10. Furthermore, in a broad sense, the rotary electric machine drive device 100 corresponds to the rotary electric machine control system, and in a narrow sense, the rotary electric machine control device 10 corresponds to the rotary electric machine control system.

[0030] The rotating motor 80 driven by the rotating motor control system is an interior permanent magnet synchronous motor (IPMSM). This IPMSM has a stator 81 with N-phase (N is any natural number) stator coils 83 arranged in a stator core 85 and a rotor 82 with permanent magnets 84 arranged in a rotor core 86. Figure 1 The illustration shows an 8-pole (4-pole pair) rotor 82 with 8 magnetic poles (4 N poles and 4 S poles), but this is illustrative and does not limit the invention. The stator 81 is similarly illustrated. Figure 1 The example illustrates a three-phase stator coil 83 short-circuited at the neutral point, but the number of phases, wiring method, and winding method of the stator coil 83 are not limited to the invention. Furthermore, the rotary motor 80 can function as both a motor and a generator. When the rotary motor 80 functions as a motor, it is in a power operation state; when it functions as a generator, it is in a regenerative state.

[0031] like Figure 1As shown, the rotary motor drive unit 100 includes a voltage-type inverter 50. The inverter 50 is connected to an AC rotary motor 80 and a DC power supply 41, converting power between multiphase AC and DC. The DC power supply 41 is, for example, composed of a rechargeable secondary battery (battery) such as a lithium-ion battery, a double-layer capacitor, etc. When the rotary motor 80 is the driving force source for the vehicle, the DC power supply 41 is a high-voltage, high-capacity DC power supply, with a rated power supply voltage of, for example, 200–400 V. On the DC side of the inverter 50, there is a smoothing capacitor (DC link capacitor 42) that smooths the voltage between the positive and negative terminals (DC link voltage Vdc).

[0032] The inverter 50 is configured with multiple switching elements 51. The switching elements 51 preferably utilize power semiconductor devices capable of operating at high frequencies, such as IGBTs (Insulated Gate Bipolar Transistors), power MOSFETs (Metal Oxide Semiconductor Field Effect Transistors), SiC-MOSFETs (Silicon Carbide-Metal Oxide Semiconductor FETs), SiC-SITs (SiC-Static Induction Transistors), and GaN-MOSFETs (Gallium Nitride MOSFETs). Figure 1 The example illustrates the use of an IGBT as a switching element 51. Furthermore, each switching element 51 has a freewheeling diode 53 connected in parallel, with the positive direction being from the negative terminal to the positive terminal (from the lower stage side to the upper stage side).

[0033] like Figure 1 As shown, the inverter 50 is controlled by the rotating motor control device 10. The rotating motor control device 10 is constructed with logic circuits such as a microcomputer as its core component. For example, the rotating motor control device 10 is based on the target torque (torque command T*: see reference) of the rotating motor 80 provided as a request signal from other control devices such as the vehicle control device 90 (VHL-CTRL), which is one of the higher-level control devices. Figure 2 The inverter 50 drives the rotating motor 80 via current feedback control using vector control. In vector control, the currents (Iu, Iv, Iw: refer to...) flowing in the stator coils 83 of each of the three (N-phase) phases of the AC motor are controlled by vector feedback. Figure 2The coordinates are converted into vector components of the magnetic field generated by the permanent magnet 84 configured on the rotor 82, i.e., the d-axis, and the q-axis, which is orthogonal to the d-axis (leading the direction of the magnetic field by π / 2 electrical angles), for feedback control. The coordinate system of the target coordinate system is called the dq-axis orthogonal coordinate system.

[0034] The actual current flowing through the stator coils 83 of each phase of the rotating electric machine 80 is detected by the current sensor 43, and the rotating electric machine control device 10 acquires the detection results. Furthermore, while the method of detecting three-phase AC current is illustrated here, in the case of three-phase AC, the three phases are balanced, and the sum of their instantaneous values ​​is zero. Therefore, it is also possible to detect the current of only two phases, with the remaining phase obtained through calculation by the rotating electric machine control device 10. Additionally, the magnetic pole position θ (electric angle) and rotational speed (angular velocity ω) of the rotor 82 of the rotating electric machine 80 at each moment are detected, for example, by a rotation sensor 44 such as a rotary transformer, and the rotating electric machine control device 10 acquires the detection results. The rotating electric machine control device 10 uses the detection results of the current sensor 43 and the rotation sensor 44 to perform current feedback control.

[0035] like Figure 2 As shown, the rotating electric machine control device 10 is configured to have various functional units for current feedback control, and each functional unit is implemented through the cooperation of hardware such as a microcomputer and software (program). In this embodiment, the rotating electric machine control device 10 includes a torque control unit (MTPA: Maximum Torque per Ampere Control) 1, a current control unit 2, a two-phase to three-phase coordinate conversion unit 3, a three-phase to two-phase coordinate conversion unit 4, a modulation unit 5 (PWM), and a correction current command setting unit 6 (Harmonic Mapping).

[0036] The torque control unit 1 sets the target current (basic current command Idq*) flowing through the stator coil 83 of the rotary motor 80 based on the torque command T* (target torque) transmitted from the vehicle control device 90. That is, the torque control unit 1 is equivalent to the basic current command setting unit. As described above, the rotary motor control device 10 performs feedback control on the rotary motor 80 in a dq-axis orthogonal vector coordinate system. Therefore, the torque control unit 1 calculates the d-axis basic current command Id* and the q-axis basic current command Iq* to obtain the basic current command Idq*. As will be described later, in this embodiment, the corrected current command Idq* obtained by superimposing the corrected current command Idqh* on the basic current command Idq* becomes the control object of the subsequent stage. That is, the corrected current command Idq* is equivalent to the target current command of the control object of the current control unit 2.

[0037] The current control unit 2 calculates the voltage command Vdq* applied to the inverter 50 based on the deviation between the corrected current command Idq** and the actual current flowing through the stator coil 83 (d-axis current Id, q-axis current Iq). The current detected by the current sensor 43 (SEN-I) is the actual current of the three phases flowing through the stator coil 83 (U-phase current Iu, V-phase current Iv, W-phase current Iw). The actual current of the three phases is converted into the current of the two phases in the dq-axis vector coordinate system (d-axis current Id, q-axis current Iq) in the three-phase two-phase coordinate transformation unit 4. The three-phase two-phase coordinate transformation unit 4 performs the coordinate transformation based on the magnetic pole position θ (electric angle) of the rotor 82 at each moment detected by the rotation sensor 44 (SEN-R).

[0038] The current control unit 2 calculates the d-axis voltage command Vd* based on the deviation between the d-axis current command (here, the corrected d-axis current command Id**) and the d-axis current Id, and the rotational speed (angular velocity ω). Similarly, it calculates the q-axis voltage command Vq* based on the deviation between the q-axis current command (here, the corrected q-axis current command Iq**) and the q-axis current Iq, and the rotational speed (angular velocity ω). Furthermore, as shown in reference... Figure 3 As will be described later, in this embodiment, the current control unit 2 is illustrated as having a proportional-integral (PI) controller, but the current control unit 2 may also be configured as a proportional-integral-derivative (PID) controller.

[0039] The two-phase to three-phase coordinate transformation unit 3 converts the two-phase voltage commands Vdq* (d-axis voltage command Vd*, q-axis voltage command Vq*) of the dq-axis vector coordinate system into three-phase voltage commands (U-phase voltage command Vu*, V-phase voltage command Vv*, W-phase voltage command Vw*) corresponding to the three-phase inverter 50. The modulation unit 5 generates three-phase switching control signals (U-phase switching control signal Su, V-phase switching control signal Sv, W-phase switching control signal Sw) for the inverter 50 based on the three-phase voltage commands (U-phase voltage command Vu*, V-phase voltage command Vv*, W-phase voltage command Vw*). Here, an example is shown where the modulation unit 5 generates the switching control signals using pulse width modulation (PWM). Furthermore, in Figure 2 In the simplified form, there are 3 switching control signals (Su, Sv, Sw), but the modulation unit 5 generates 6 switching control signals (U-phase upper-level switching control signal, U-phase lower-level switching control signal, ...) corresponding to the 6 switching elements 51 of the inverter 50.

[0040] like Figure 1 and Figure 2As shown, the control terminals (e.g., the gate terminals of IGBTs) of each switching element 51 constituting the inverter 50 are connected to the rotary motor control device 10 via the drive circuit 15 (DRV-CCT), and each switching element 51 is independently controlled for switching. As described above, the rotary motor control device 10, which generates the switching control signals, is configured with a microcomputer or similar core, and its operating voltage is, for example, 5V or 3.3V. On the other hand, the inverter 50, as described above, is connected to a DC power supply 41 with a rated power supply voltage of, for example, 200V to 400V, and requires a drive signal of, for example, 15V to 20V, to be input to the control terminals of the switching elements 51. The drive circuit 15 enhances the driving capability (e.g., the ability of voltage amplitude, output current, etc., to operate subsequent circuits) of the switching control signals generated by the rotary motor control device 10 and relays it to the inverter 50.

[0041] However, in the rotary motor 80 with embedded permanent magnets, the flux linkage changes as the rotor 82 rotates, resulting in torque fluctuations. That is, in Figure 1 An excitation force (torque fluctuation) is generated along the circumferential direction C shown. Additionally, due to the attraction and rebound force between the stator core 85 and the permanent magnet 84, in... Figure 1 The torque on the radial R also varies. The excitation force of this radial R is a radial excitation force. When the rotor 82 vibrates due to these excitation forces, audible noise may sometimes be produced. This audible noise can sometimes be harsh to the user, so it is preferable to reduce these excitation forces. Regarding the audible noise, the torque fluctuation, as a circumferential excitation force, contributes significantly. In this embodiment, torque fluctuation is reduced by generating a torque that eliminates torque fluctuation. The aforementioned corrected current command Idq** is a command obtained by superimposing the corrected current command Idqh*, which generates the torque for eliminating torque fluctuation, with the basic current command Idq*.

[0042] like Figure 2 As shown, the rotary motor control device 10 of this embodiment includes a correction current command setting unit 6, which sets a correction current command Idqh* superimposed on the basic current command Idq* to reduce torque fluctuations of the rotary motor 80. The correction current command setting unit 6 is configured to have a mapping (higher harmonic mapping) with torque command T* and magnetic pole position θ as arguments.

[0043] As described above, the d-axis correction current command Idh* and the q-axis correction current command Iqh* are superimposed on the d-axis basic current command Id* and the q-axis basic current command Iq* set by the torque control unit 1, respectively. The current control 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, as well as the rotational speed (angular velocity ω), and calculates the q-axis voltage command Vq* based on the deviation between the corrected q-axis current command Iq** and the q-axis current Iq, as well as the rotational speed (angular velocity ω). As a result, the rotary motor 80 can output torque with reduced torque fluctuations.

[0044] The correction torque used to reduce excitation force is from Figure 4 The torque in the opposite phase of the (2NM)th higher harmonic torque component (M is any natural number) of the circumferential excitation force within the higher harmonic torque component (torque fluctuation) extracted from the actual torque T of the rotary electric machine 80 is shown. In this embodiment, N, representing the number of phases of the alternating current, is 3. Therefore, for example, when set to "M=1", the torque in the opposite phase of the 6th higher harmonic torque component within the higher harmonic torque component extracted from the actual torque T becomes the correction torque. Figure 5 The solid line represents the waveform from Figure 4 The actual torque T shown is the (2NM)th higher harmonic torque component (torque fluctuation) extracted. The waveform of the dotted line represents the correction torque of the (2NM)th higher harmonic torque component in the opposite phase.

[0045] The phase of the correction current command Idqh* is set so that the phase of the correction torque differs from the phase of the torque fluctuation by 180 degrees. Additionally, in Figure 5 In the example shown, the amplitude of the torque ripple is equal to the amplitude of the corrective torque. Thus, when the phase of the torque ripple is 180 degrees out of phase with the corrective torque and the amplitudes are equal, the corrective torque and the torque ripple cancel each other out, and almost all the torque ripple is reduced.

[0046] Here, as Figure 5 As shown, the peak value of the side with the larger correction torque is designated as the first torque T1, the peak value of the side with the smaller correction torque is designated as the second torque T2, and the average value of the correction torque is designated as the average correction torque Tav. The average correction torque Tav corresponds to the correction torque and the DC component of the actual torque T of the rotary motor 80. When the current feedback control based on the torque command T* is properly executed, the output torque of the rotary motor 80 becomes approximately equal to the torque command T* (target torque), unaffected by the correction torque.

[0047] However, as described above, the basic current command Idq* is a direct current, but the corrected current command Idqh* is an alternating current. Therefore, the corrected current command Idq* has both a direct current component and an alternating current component. A typical current control unit 2 is configured with a proportional-integral (PI) controller, but the responsiveness of a PI controller to high frequencies is limited. That is, the current control may not be able to adequately track the alternating current component corresponding to the corrected current command Idqh*. Therefore, in this embodiment, the current control unit 2 includes: a first current control unit 20, which performs current control on the direct current component corresponding to the basic current command Idq*; and a second current control unit 21, which performs current control on the alternating current component corresponding to the corrected current command Idqh*.

[0048] The first current control unit 20, like a general current control unit, is composed of a proportional-integral (PI) controller.

[0049] The second current control unit 21 converts the AC component coordinates, equivalent to the correction current command Idqh*, into DC components and performs proportional-integral control. By converting the DC component inverse coordinates into AC components, it performs current control on the correction current command Idqh*. Figure 3 As shown, the second current control unit 21 performs coordinate transformation on the AC component corresponding to the correction current command Idqh* from the dq-axis orthogonal vector coordinate system to the γδ-axis orthogonal coordinate system. After passing through a low-pass filter, it performs proportional-integral control. By performing an inverse coordinate transformation from the γδ-axis orthogonal coordinate system to the dq-axis orthogonal vector coordinate system, it performs current control on the correction current command Idqh*. The second current control unit 21 has two systems (paths 22-25 and 26-29) to correspond to the cases where the phase of the dq-axis current Idq leads and lags the correction current command Idqh, respectively. Reference numerals "22" and "26" are coordinate transformation units, reference numerals "23" and "27" are low-pass filters (LPF), reference numerals "24" and "28" are proportional-integral controllers (PI), and reference numerals "25" and "29" are inverse coordinate transformation units.

[0050] Figure 6 This shows the relationship between torque and basic current command Idq* in the orthogonal coordinate system of the dq axis. Figure 6The curve indicated by reference numeral "30" in the attached figure is the isotorque line representing the combination of d-axis and q-axis currents (the vector trajectory of the currents in the dq orthogonal vector coordinate system) capable of outputting constant torque. Reference numeral "31" is the isotorque line 30 for the first torque T1, i.e., the first isotorque line, and reference numeral "32" is the isotorque line 30 for the second torque T2, i.e., the second isotorque line. Furthermore, reference numeral "33" is the isotorque line 30 for the reference torque T0, i.e., the reference isotorque line, which corresponds to the target torque (torque command T*) of the aforementioned rotary motor 80.

[0051] The reference numeral "60" in the attached diagram represents the basic control line (the vector trajectory of the current in the d-axis current in the dq orthogonal vector coordinate system) when the rotary electric machine 80 is controlled under standard conditions (hereinafter referred to as "basic control"). Generally, the basic control line 60 represents the vector trajectory of the optimal combination of d-axis and q-axis currents for outputting arbitrary torque in the dq orthogonal vector coordinate system. As an example, the basic control line 60 may be the maximum torque line or the maximum efficiency line, representing the vector trajectory of the combination of d-axis and q-axis currents that can output each torque with the highest efficiency. The torque control unit 1 may be a functional unit that sets the basic current command Idq* representing such a vector trajectory.

[0052] For example, without considering the reduction in torque fluctuation, that is, when simply setting the d-axis basic current command Id* and the q-axis basic current command Iq* based on the torque command T*, setting... Figure 6 The current value at reference point P0 is shown. Reference point P0 is the intersection of the basic control line 60 and the isotorque line (reference isotorque line 33) corresponding to the torque command T* (target torque) in the dq-axis orthogonal vector coordinate system. In this embodiment, to reduce torque fluctuations, as shown in the reference... Figure 5 The aforementioned method superimposes the correction current command Idqh*, which is capable of outputting correction torque, with the d-axis basic current command Id* and the q-axis basic current command Iq*. In other words, the d-axis correction current command Idh* and the q-axis correction current command Iqh*, which are composed of AC components (here, (2NM) higher harmonic components), are superimposed with the d-axis basic current command Id* and the q-axis basic current command Iq* (the current command Idq* corresponding to the torque command T*), which are DC components in the dq-axis orthogonal vector coordinate system.

[0053] For example, the correction torque is the torque that oscillates between the first torque T1 and the second torque T2 via the average correction torque Tav (reference torque T0). The vector trajectories of the d-axis correction current command Idh* and the q-axis correction current command Iqh* used to output the correction torque in the dq-axis orthogonal vector coordinate system are straight lines (segments) that connect, for example, the first equal torque line 31 and the second equal torque line 32 through the reference point P0. Hereinafter, this straight line passing through the reference point P0 will be referred to as the correction line K. In addition, the intersection of the first equal torque line 31 and the correction line K is referred to as the first intersection point P1, and the intersection of the second equal torque line 32 and the correction line K is referred to as the second intersection point P2.

[0054] The correction line K can theoretically be set to infinity. Figure 6 Three calibration lines K (K11, K12, and K13) are illustrated. The first calibration line K11 represents the way torque changes by varying the current along the q-axis, the second calibration line K12 represents the way torque changes by varying the current along the d-axis, and the third calibration line K13 represents the way torque changes by varying the current in a direction inclined relative to the d-axis and q-axis.

[0055] When the vector trajectory of the correction current command Idqh* is the first correction line K11, the correction current command Idqh* consists only of the q-axis correction current command Iqh*. Since the first correction line K11 is parallel to the q-axis, the d-axis correction current command Idh* is a constant value, which is the value of the d-axis basic current command Id* corresponding to the torque command T* (the value of the d-axis current at the reference point P0). When the vector trajectory of the correction current command Idqh* is the second correction line K12, the correction current command Idqh* consists only of the d-axis correction current command Idh*. Since the second correction line K12 is parallel to the d-axis, the q-axis correction current command Iqh* is a constant value, which is the value of the q-axis current at the reference point P0. When the vector trajectory of the correction current command Idqh* becomes the third correction line K13, the correction current command Idqh* consists of both the d-axis correction current command Idh* and the q-axis correction current command Iqh*.

[0056] The correction current command Idqh* can set multiple (2NM) order higher harmonic torque components as targets for reduction. Furthermore, the correction line K can vary depending on the order of the higher harmonic torque components. In this embodiment, the 6th and 12th order higher harmonic torque components are targeted for reduction. A first correction line K11 is set for the 6th order higher harmonic torque component, and a third correction line K13 is set for the 12th order higher harmonic component.

[0057] In addition, such as Figure 7 and Figure 8As shown, the amplitude and phase of the correction current command Idqh* vary depending on the torque (torque command T*) of the rotating electric machine 80. Figure 7 The graph shows the relationship between the amplitude of the correction current command Idqh* and the torque (torque command T*). Figure 8 The diagram illustrates the relationship between the phase of the correction current command Idqh* and the torque (torque command T*). In the diagram, "fe" represents the frequency of the electrical angle, "6fe" represents the 6th harmonic, and "12fe" represents the 12th harmonic. Additionally, "d" in parentheses represents the d-axis correction current command Idh*, and "q" in parentheses represents the q-axis correction current command Iqh*. Thus, the correction current command Idqh* is the frequency corresponding to the torque fluctuations of the object being reduced, and it is an alternating current whose phase and amplitude differ from the torque command T* of the rotating motor 80.

[0058] However, a correction torque is output to suppress vibrations caused by torque fluctuations to the point of being audible. The frequency of torque fluctuations varies depending on the rotational speed of the rotary motor 80. Furthermore, the frequency at which the vibrations become audible is approximately 20 Hz to 20 kHz, and frequencies below 1 kHz are particularly harsh to the human ear. Therefore, in this embodiment, the correction current command setting unit 6 sets the correction current command Idqh* when the rotational speed of the rotary motor 80 is at a predetermined target rotational speed. Figure 9 The operating range of the rotary motor 80 based on torque and speed is shown. Within this operating range, a correction speed ST is set between the first target speed S1 and the second target speed S2.

[0059] The frequency of the vibration caused by torque fluctuation is determined by the following equation (1) based on the structure of rotor 82 (number of pole pairs), the rotational speed of rotor 82, and the frequency of the higher harmonics of torque fluctuation. In this embodiment, the rotor 82 of the rotary motor 80 has 4 pole pairs. Here, when the rotational speed of the rotary motor 80 is 1000 rpm, the frequency of the vibration caused by the 6th higher harmonic is 400 Hz as shown in the following equation (2), and the frequency of the vibration caused by the 12th higher harmonic is 800 Hz as shown in the following equation (3).

[0060] Vibration frequency [Hz] = Rotational speed [rpm]·(Number of pole pairs / 60 [sec])·2MN···(1)

[0061] 400[Hz]=1000[rpm]·(4 / 60[sec])·6···(2)

[0062] 800[Hz]=1000[rpm]·(4 / 60[sec])·12···(3)

[0063] Therefore, in the case of a rotary motor 80 having a 4-pole pair rotor 82 as in this embodiment, the target speed ST is set to 1000 rpm.

[0064] Furthermore, when the torque is small, torque fluctuation is also small, resulting in less audible noise. Conversely, when the torque is large, the impact of audible noise caused by torque fluctuation is relatively smaller. Therefore, when the torque (torque command T*) of the rotary motor 80 is a predetermined torque to be calibrated, the calibration current command setting unit 6 sets the calibration current command Idqh*. In this embodiment, when... Figure 9 In the shown operating range, when the absolute value of the torque of the target torque is between "a / 2" and "e", the correction current command Idqh* is set. For example... Figure 9 As shown, in this embodiment, in the calibration target action region H set according to the rotational speed and torque (torque command T*) of the rotary motor 80, a calibration current command Idqh* is set to output the calibration torque.

[0065] In this way, by not superimposing the correction current command Idqh* in the entire operating region of the rotating motor 80, the efficiency reduction caused by the superposition of the correction current command Idqh* can be suppressed. Furthermore, in the operating region where torque fluctuations need to be suppressed, by superimposing the correction current command Idqh*, torque fluctuations can be appropriately suppressed.

[0066] However, refer to Figure 8 It can be seen that when the torque of the rotating motor 80 varies between "b" and "c", the phase change of the correction current command Idqh* (d-axis correction current command Idh* of the 12th harmonic) is larger compared to the torque changes in other parts. The phase of the correction current command Idqh* is set so that the torque output based on the correction current command Idqh* (corrected torque) becomes the opposite phase of the torque fluctuation. That is, as referenced... Figure 5 The phase of the correction current command Idqh* is set such that the phase of the correction torque differs from the phase of the torque ripple by 180 degrees. Therefore, if the phase of the correction current command Idqh* differs from the appropriate phase, the phase difference between the correction torque and the torque ripple shifts by 180 degrees, reducing the torque ripple reduction effect. For example, if the phase of the correction current command Idqh* shifts by 180 degrees from the appropriate phase, the phase of the correction torque coincides with the phase of the torque ripple, potentially causing the torque ripple to increase by up to twice, thus exciting the rotating motor 80.

[0067] like Figure 8As shown, the d-axis correction current command Idh* of the 12th harmonic changes phase by more than 180 degrees when the torque varies between "b" and "c". For example, when the torque of the rotary motor 80 increases from "b" to "c" during vehicle acceleration, the phase change of the d-axis correction current command Idh* of the 12th harmonic exceeds 180 degrees. Here, if the delay from the start of the torque command T* change to the setting of the correction current command Idqh* by the correction current command setting unit 6, and the response delay in the current control unit 2, increase, the correction torque based on the correction current command Idqh* may increase torque fluctuations and apply excitation to the rotary motor 80.

[0068] Therefore, in this embodiment, in the operating region where the phase of the correction current command Idqh* changes drastically relative to the torque, particularly in the operating region corresponding to the inflection point where the phase reverses relative to the torque, the control to reduce torque fluctuations based on the correction torque is limited. The correction current command setting unit 6, in a specific operating region E (referring to...) that constitutes such an operating region... Figure 9 In this process, the amplitude of the correction current command Idqh* is reduced. For example, the correction current command setting unit 6 can also substantially disable correction by setting the amplitude of the correction current command Idqh* to zero.

[0069] The specific action range E is set based on the torque command T*. For example, when... Figure 9 The torque (torque command T*) shown is set as a specific action area E for the action area of ​​“b” to “c” and the action area of ​​“-b” to “-c”. Figure 9 This example illustrates a method of determining that the absolute values ​​of the torque in the first specific operating region E1 during power operation and the second specific operating region E2 during regeneration are the same, but the absolute values ​​of the torque during power operation and regeneration can also be different. Furthermore, it is not excluded that the specific operating region E may be set only during either power operation or regeneration, for example, only during power operation.

[0070] Furthermore, in this embodiment, the specific operating region E is set as an operating region where the change in phase of the correction current command Idqh* relative to the change in torque command T* is greater than or equal to a predetermined value. As described above, when the change in phase increases, the difference between the phase of the torque fluctuation of the target being reduced and the phase of the correction torque is less than 180 degrees, and the torque fluctuation reduction effect decreases. The more abrupt the change in phase of the correction current command Idqh*, the easier it is for the phase of the correction torque to shift relative to the phase of the torque fluctuation, and the larger this phase shift becomes. In other words, the greater the change in phase of the correction current command Idqh* relative to the change in torque command T*, the lower the torque fluctuation reduction effect; conversely, torque fluctuation may sometimes increase.

[0071] Therefore, preferably, the specific operating region E is set to at least a point where torque fluctuation does not increase. When the change in the phase of the correction current command Idqh* relative to the change in torque command T* is relatively small, even if the tracking of the change in correction current command Idqh* is delayed, the phase shift of the correction torque relative to the phase of torque fluctuation is small. On the other hand, when the change in the phase of the correction current command Idqh* relative to the change in torque command T* is relatively large, when the tracking of the change in correction current command Idqh* is delayed, the phase shift of the correction torque relative to the phase of torque fluctuation also becomes larger, resulting in a decrease in the torque fluctuation reduction effect and a higher probability of an increase in torque fluctuation. For example, if the change in the phase of the correction current command Idqh* relative to the change in torque command T* by "1" is set to a value of about 5 to 10 degrees, when the change is greater than or equal to this value, limiting the superposition of correction current command Idqh* can suppress the amplification of torque fluctuation, and when the torque fluctuation suppression effect is low, it can suppress the reduction in control efficiency of the rotary motor 80 due to the superposition of correction current command Idqh*.

[0072] As described above, the limitation on superimposing the correction current command Idqh* onto the basic current command Idq* to avoid generating correction torque is due to the responsiveness of the correction current command setting unit 6 and the current control unit 2. That is, it is due to the convergence time relative to the change in (target torque) of the torque command T*, which is the time it takes for the phase of the correction current command Idqh* to track to the phase corresponding to the changed torque command T*. Therefore, preferably, a specific operating region E is set based on the operating region where the correction current command Idqh* amplifies the torque fluctuation (torque vibration) when the maximum possible torque change occurs within the convergence time.

[0073] If a specific motion region E is set in this way, torque fluctuations can be appropriately reduced in motion regions that the control can track, and torque fluctuations can be amplified without correcting torque in motion regions that the control cannot track.

[0074] Figure 10 The graph illustrates the relationship between the reduction effect of torque ripple and the phase difference between torque ripple and corrective torque. Furthermore, equations (4) and (5) below schematically represent torque ripple and corrective torque. Equation (4) represents torque ripple "F", and equation (5) represents corrective torque "G". In addition, "π" in equation (5) is 180 degrees, and when "α" and "β" are the same, it indicates that torque ripple and corrective torque are out of phase (phase difference of 180 degrees).

[0075] F=Acos(ωt+α)···(4)

[0076] G=Bcos(ωt+β+π)···(5)

[0077] exist Figure 10 In the chart, the vertical axis represents the absolute value (|FG|) of the torque fluctuation "F" minus the corrective torque "G", and the horizontal axis represents the phase difference (β - α) between the torque fluctuation "F" and the corrective torque "G". Additionally, in Figure 10 The diagram shows two characteristic curves. One is for the case where the amplitudes of the torque ripple "F" and the corrective torque "G" are the same (A = B), and the amplitude ratio (B / A) is "1". The other is for the case where the amplitude of the corrective torque "G" relative to the torque ripple "F" is "1 / 2" (B = A / 2), and the amplitude ratio (B / A) is "0.5".

[0078] When the amplitude "A" of the torque ripple (torque vibration) reduction is the same as the amplitude "B" of the correction torque based on the correction current command Idqh*, and "α" and "β" are consistent, the vectors of torque ripple and correction torque have the same magnitude and are in opposite directions, resulting in a composite vector of zero magnitude. The magnitude of this composite vector represents the magnitude of the residual torque ripple after applying the correction torque. When "α" and "β" are consistent, the reduction effect of torque ripple based on the correction torque is maximized, effectively offsetting the torque ripple through the correction torque.

[0079] When there is a difference between "α" and "β", the magnitude of the resultant vector increases as the difference increases. That is, the magnitude of the residual torque fluctuation increases. When the difference between "α" and "β" is ±60 degrees, the magnitude of the resultant vector is the same as the original torque fluctuation. In other words, when the difference between "α" and "β" is ±60 degrees, the torque fluctuation reduction effect disappears. When the difference between "α" and "β" is greater than ±60 degrees, the magnitude of the resultant vector exceeds the original torque fluctuation. That is, by applying a corrective torque, the torque fluctuation is amplified. Therefore, as... Figure 10 As shown, when the amplitude of the torque fluctuation is the same as the amplitude of the correction torque, the correctable range D of the torque fluctuation is the first correctable range D1, where the difference between "α" and "β" is within ±60 degrees.

[0080] On the other hand, when the amplitude of the corrective torque is "1 / 2" relative to the amplitude of the torque fluctuation and "α" and "β" are the same, the magnitude of the vector of the corrective torque is "1 / 2" relative to the vector of the torque fluctuation and they are in opposite directions, and the magnitude of the resultant vector is "1 / 2" of the magnitude of the torque fluctuation. That is, when the amplitude of the corrective torque is "1 / 2" of the torque fluctuation, even if the reduction effect of the torque fluctuation caused by the corrective torque is maximized, the torque fluctuation cannot be completely offset by the corrective torque; even with the corrective torque applied, the torque fluctuation remains. In other words, when the amplitude of the corrective torque is "1 / 2" relative to the amplitude of the torque fluctuation, the reduction effect of the torque fluctuation is lower compared to the case where the amplitude of the torque fluctuation and the amplitude of the corrective torque are the same. However, as will be discussed later, compared to the case where the amplitude of the torque fluctuation and the amplitude of the corrective torque are the same, the correctable range D of the torque fluctuation becomes larger.

[0081] When a difference exists between "α" and "β", similar to the case where the amplitude of the torque fluctuation and the amplitude of the corrective torque are the same, the magnitude of the resultant vector increases as this difference increases. That is, the magnitude of the residual torque fluctuation increases. When the difference between "α" and "β" is ±60 degrees, the magnitude of the resultant vector is approximately 0.87 (=(3 1 / 2 (2) When the difference between "α" and "β" is approximately ±80 degrees, the reduction effect of torque ripple disappears. When the difference between "α" and "β" is greater than the aforementioned ±80 degrees, the magnitude of the composite vector exceeds the original magnitude of torque ripple. In other words, by applying a correction torque, torque ripple is amplified.

[0082] like Figure 10 As shown, when the amplitude of the corrected torque is "1 / 2" relative to the amplitude of the torque fluctuation, the correctable range D of the torque fluctuation is a second correctable range D2, where the difference between "α" and "β" is approximately ±80 degrees. The second correctable range D2 is wider than the first correctable range D1. Thus, when the amplitude of the corrected torque is small relative to the amplitude of the torque fluctuation, the reduction effect of torque fluctuation is lower compared to the case where the amplitudes of the torque fluctuation and the corrected torque are the same, but even with a phase shift, the range for obtaining the reduction effect of torque fluctuation is larger.

[0083] In the case of reducing the priority torque fluctuation, as a method, preferably, the amplitude of the reduced torque fluctuation is the same as the amplitude of the corrected torque based on the corrected current command Idqh*, and the specified value of the change in the phase of the corrected current command Idqh* is ±60 degrees.

[0084] The maximum reduction effect can be achieved if the amplitude "A" of the torque fluctuation reduction is the same as the amplitude "B" of the corrected torque based on the corrected current command Idqh*. However, with the optimal phase offset of the corrected torque, the reduction effect decreases as the offset increases, and the torque fluctuation is amplified when the phase offset exceeds 60 degrees. Therefore, if a predetermined value is set to 60 degrees, and the action region where the phase change of the corrected current command Idqh* relative to the change in torque command T* (target torque) is above a specified value is set as a specific action region E, then even if the torque command T* (target torque) changes drastically, the torque fluctuation will not be amplified, thereby appropriately reducing the torque fluctuation.

[0085] Furthermore, as described above, since the torque ripple reduction effect disappears at ±60 degrees, it is preferable to set the predetermined value to a range narrower than ±60 degrees when the amplitude of the torque ripple of the object being reduced is the same as the amplitude of the corrected torque based on the corrected current command Idqh*. In this case, since the corrected current command Idqh* is not superimposed in the operating region where the torque ripple reduction effect is low, the rotary motor control device 10 can control the rotary motor 80 with high efficiency.

[0086] If the above is organized, the specific motion region E can be set according to the following conditions. Assume the motion point in the orthogonal coordinate system of the dq axes originates from the torque T before movement. bfr [Nm] Torque T after movement aft [Nm]. Here, the torque response rate is set to Ta[Nm / s], and the convergence time of the correction current command Idqh* is set to t[ms]. The torque that can vary during the convergence time "t[ms]" is "Ta·t", and the torque T before movement is... bfr [Nm] and torque after movement T aft The differential torque "ΔT[Nm]" is equivalent to this. Additionally, the torque T before movement... bfr The phase of the correction current command Idqh* in r[Nm] is set to "θ bfr "The torque T after the movement" aft The phase of the correction current command Idqh* in [Nm] is set to "θ". aft The specific action area E is set to "θ". bfr "and "θ aft The absolute value of the difference between "|θ" aft -θ bfr | "Exceeding the permissible phase difference" The range of torque. Furthermore, the permissible phase difference. It varies according to the ratio of the amplitude of the torque fluctuation to the amplitude of the corrective torque. (See reference...) Figure 10 The statement states that when the ratio is "1", it is " Spend".

[0087] However, as mentioned above Figure 3 In order to improve the responsiveness of current control to higher harmonic components, i.e., the correction current command Idqh*, the current control unit 2 has a second current control unit 21. Figure 11 The graph shows the frequency characteristics of the current control unit 2, with the vertical axis representing gain and the horizontal axis representing frequency. Furthermore, Figure 11 It is a logarithmic curve with the horizontal axis as the logarithmic axis. Figure 13 and Figure 14 (The same applies). In Figure 11 In the diagram, the double-dotted line represents the frequency characteristic of only the first current control unit 20, while the solid line represents the overall frequency characteristic of the current control unit 2 after merging the first current control unit 20 and the second current control unit 21. Furthermore, (6fe) in the diagram represents the frequency corresponding to the correction current command Idqh* for the 6th harmonic, and (12fe) represents the frequency corresponding to the correction current command Idqh* for the 12th harmonic.

[0088] As described above, the second current control unit 21 is provided to improve the responsiveness of proportional-integral control for alternating current including higher harmonic components, i.e., the correction current command Idqh*. Figure 11 As shown, in the frequency characteristics of only the first current control unit 20, the gain at the frequencies corresponding to the correction current command Idqh* of the 6th higher harmonic and the correction current command Idqh* of the 12th higher harmonic is low, and the current control response to the correction current command Idqh* is insufficient.

[0089] In contrast, the overall frequency characteristics of the current control unit 2, which combines the first current control unit 20 and the second current control unit 21, show that the gain at frequencies corresponding to the 6th harmonic correction current command Idqh* and the 12th harmonic correction current command Idqh* does not decrease, indicating sufficient responsiveness to the current control of the correction current command Idqh*. However, at frequencies higher than those corresponding to the 12th harmonic correction current command Idqh*, the gain increases significantly. For example, as... Figure 12 As shown, when the torque command T* increases sharply like a step response, the basic current command Idq* also exhibits a step response. In the current control unit 2, since high-frequency control corresponding to the increase in the step response is executed, therefore, as... Figure 11 This produces a large gain. The result is as follows: Figure 12 As shown, overshoot may occur in the output torque, or vibration may occur afterward.

[0090] Therefore, in this embodiment, the gain of the second current control unit 21 is set to be lower than the gain of the first current control unit 20. Figure 13 and Figure 14 The frequency characteristics are illustrated when the gain of the second current control unit 21 is set to "1 / 6" of the gain of the first current control unit 20. Figure 13 The rotational speed of the rotary motor 80 is shown at... Figure 9 The frequency characteristics near the first object rotational speed S1 Figure 14 The rotational speed of the rotary motor 80 is shown at... Figure 9 The frequency characteristic between the first object rotational speed S1 and the second object rotational speed S2. That is, relative to Figure 13 , Figure 14 The frequency characteristics of the rotary motor 80 at high speeds are shown. The frequency corresponding to the correction current command Idqh* also depends on the rotational speed of the rotary motor 80 (refer to equations (1) to (3)). Therefore, with Figure 13 compared to, Figure 14 The frequencies corresponding to the correction current commands Idqh* for the 6th and 12th harmonics also shift towards higher frequencies.

[0091] like Figure 13 and Figure 14 As shown, the second current control unit 21 responds appropriately at the frequency controlled by the second current control unit 21, suppressing excessive responses at other frequencies. Therefore, even when the torque command T* (target torque) changes drastically, the output torque will not overshoot or subsequently vibrate, thus enabling proper control of the rotary motor 80.

[0092] Thus, when the gain of the second current control unit 21 is lower than that of the first current control unit 20, the responsiveness of the current control to the correction current command Idqh* decreases, and therefore, the convergence time may become longer. Therefore, in this embodiment, a correction voltage command setting unit 7 (Harmonic Voltage Map) is also provided. This unit sets a correction voltage command Vdqh* superimposed on the voltage command Vdq* using feedforward control to reduce torque fluctuations in the rotary motor 80. Like the correction current command setting unit 6, the correction voltage command setting unit 7 is composed of a mapping. This mapping is set based on a stable value obtained by superimposing the correction current command Idqh* and the basic current command Idq* through experiments or simulations and performing current control to achieve convergence.

[0093] By using feedforward control to superimpose the correction voltage command Vdqh* with the voltage command Vdq*, the convergence time of the current control unit 2 is shortened, and the responsiveness is improved. For example, even if the gain of the second current control unit 21 is set lower than that of the first current control unit 20, the voltage command Vdq*, which includes the correction voltage command Vdqh* used to generate the correction torque, can be calculated appropriately.

[0094] Furthermore, the correction voltage command setting unit 7 sets the correction voltage command Vdqh* based on the torque command T*, the rotational speed of the rotary motor 80, and the magnetic pole position θ. Torque fluctuations occur based on the output torque and rotational speed of the rotary motor 80. Therefore, the correction voltage command Vdqh* is a voltage command used to output a correction torque to reduce torque fluctuations, and thus it is appropriately set based on the torque command T* and the rotational speed of the rotary motor 80.

[0095] Figure 15 This is a waveform diagram showing an example of the experimental or simulation results of the dq-axis current Idq in the case of a corrected current command Idqh* without superimposed higher harmonics. Figure 16 This is a waveform diagram showing an example of the experimental or simulation results of the dq-axis current Idq in the case of a corrected current command Idqh* with superimposed higher harmonics. Figure 15 and Figure 16 In the diagram, the upper waveform represents the d-axis current Id, and the lower waveform represents the q-axis current Iq. Additionally, in... Figure 15 and Figure 16 In the image, the waveform on the left shows the torque command T* changing from zero to "e[Nm]" (see reference). Figure 7 , Figure 8 The waveform diagram on the right shows the timing when the torque command T* changes from "e[Nm]" to zero.

[0096] Any waveform represents the dq-axis current waveform under the condition that the torque command T* changes in a stepwise manner. No overshoot or vibration was observed in any waveform. In other words, it was confirmed that even when the correction current command Idqh* is superimposed on the basic current command Idq*, the interference of the current control unit 2 can be suppressed in the same way as when it is not superimposed.

[0097] [Other Implementation Methods]

[0098] The following describes other implementation methods. Furthermore, the structures of the various implementation methods described below are not limited to individual application; they can be combined with the structures of other implementation methods as long as no contradiction arises.

[0099] (1) In the above, it was illustrated that the current control unit 2 has a first current control unit 20 and a second current control unit 21, and that in addition to the current control unit 2 which performs feedback control, it also has a correction voltage command setting unit 7 which performs feedforward control. However, it is not ruled out that the current control unit 2 may only have the first current control unit 20 and may not have the correction voltage command setting unit 7. In addition, the current control unit 2 may have the correction voltage command setting unit 7 even if it does not have the second current control unit 21, and the current control unit 2 may have the second current control unit 21 even if it does not have the correction voltage command setting unit 7. Moreover, in these cases, it is preferable that when the change in the phase of the correction current command Idqh* is large relative to the change in the torque command T* of the rotating motor 80, a specific operating region E is set to limit the superposition of the correction current command Idqh*.

[0100] (2) In the above description, an example is given of setting a portion of the entire operating region of the rotary motor 80 as the target of torque fluctuation reduction, i.e., the correction target operating region H (see reference). Figure 9 However, the entire operating area of ​​the rotary motor 80 can also be the operating area H of the correction target.

[0101] (3) As described above, the current control unit 2 for performing feedback control includes a first current control unit 20 and a second current control unit 21, with the gain of the second current control unit 21 set lower than the gain of the first current control unit 20, and also includes a correction voltage command setting unit 7 for performing feedforward control. However, if the convergence time of the control can be satisfied, the rotary motor control device 10 may be configured without the correction voltage command setting unit 7. Furthermore, in the configuration where the gain of the second current control unit 21 is set lower than the gain of the first current control unit 20, the gain of the second current control unit 21 may be set to zero. In this case, the current control unit 2 is essentially composed only of the first current control unit 20. Moreover, in this case, in addition to the first current control unit 20 (current control unit 2), it is preferable to also include the correction voltage command setting unit 7.

[0102] [Summary of Implementation Methods]

[0103] The following is a brief overview of the rotary electric machine control device 10 (rotary electric machine control system) described above.

[0104] As one approach, a rotating motor control system (100) uses a permanent magnet type rotating motor (80) driven by N-phase AC (N is any natural number) as the controlled object, and performs current feedback control in a dq-axis orthogonal vector coordinate system along the d-axis of the direction of the excitation flux generated by the permanent magnet (84) and the q-axis orthogonal to the d-axis. The rotating motor control system includes: a basic current command setting unit (1) that sets a basic current command (Idq*) as a command value of the current flowing in the rotating motor (84), i.e., a current command, based on the target torque (T*) of the rotating motor (80); and a correction current command setting unit (6) that sets a correction current command (Idqh*), which is superimposed on the basic current command (Idq*) to reduce torque fluctuations of the rotating motor (80). The current command is an alternating current with a frequency corresponding to the torque fluctuation of the object being reduced; and the current control unit (2) performs current feedback control based on the deviation between the object current command (Idq**) of the controlled object obtained by adding the basic current command (Idq*) and the correction current command (Idqh*) and the current flowing through the rotating motor (80), i.e., the actual current, and calculates the command value of the voltage applied to the rotating motor (80), i.e., the voltage command (Vdq*). The current control unit (2) has a first current control unit (20) that takes the component of the basic current command (Idq*) as the controlled object and a second current control unit (21) that takes the component of the correction current command (Idq**) as the controlled object. The gain of the second current control unit (21) is set to be lower than the gain of the first current control unit (20).

[0105] The component of the basic current command (Idq*) controlled by the first current control unit (20) is a direct current in the orthogonal vector coordinate system of the dq axis, while the component of the correction current command (Idqh*) controlled by the second current control unit (21) is an alternating current in the orthogonal vector coordinate system of the dq axis. By having a first current control unit (20) and a second current control unit (21) with different control objects, the basic current command (Idq*) superimposed with the correction current command (Idqh*) can be used as the control object to perform appropriate current control. On the other hand, when the target torque (T*) changes drastically, the basic current command (Idq*) also changes drastically in response. Since this point of change contains high-frequency components, the second current control unit (21) also performs current control targeting the basic current command (Idq*), which may result in overshoot or vibration in the output voltage command (Vdq*). In other words, control interference may occur. However, according to this structure, since the gain of the second current (21) is set lower than the gain of the first current control unit (20), such overshoot is suppressed, and the generation of vibration is also suppressed. That is, according to this structure, even in a rotating motor with torque variation, no control interference is generated, thereby appropriately reducing torque fluctuation.

[0106] In addition, preferably, the rotary motor control system (100) sets a correction voltage command (Vdqh*) that is superimposed on the voltage command (Vdq*) using feedforward control to reduce torque fluctuations of the rotary motor (80).

[0107] By superimposing a correction voltage command (Vdqh*) onto the voltage command (Vdq*) using feedforward control, the convergence time of the current control unit (2) is shortened, and the responsiveness is improved. For example, even if the gain of the second current control unit (21) is set lower than that of the first current control unit (20), the voltage command (Vdq*) including the voltage command (Vdqh*) of the correction voltage command setting unit (7) for generating correction torque can be appropriately calculated.

[0108] Furthermore, in the case where a correction voltage command setting unit (7) is provided as described above, preferably, the correction voltage command setting unit (7) sets the correction voltage command (Vdqh*) based on the target torque (T*), the rotational speed of the rotary motor (80), and the magnetic pole position (θ) of the rotary motor (80).

[0109] Torque fluctuations arise based on the output torque and speed of the rotating motor 80. Therefore, the correction voltage command (Vdqh*) is a voltage command (Vdq*) used to output a correction torque to reduce torque fluctuations. Additionally, the magnetic pole position (θ) is required for coordinate transformation. Therefore, preferably, the correction voltage command (Vdqh*) is appropriately set based on the torque command (T*), the speed of the rotating motor (80), and the magnetic pole position (θ).

[0110] Furthermore, preferably, the correction current command (Idqh*) is an alternating current whose phase differs from the target torque (T*) of the rotating motor (80). The correction current command setting unit (6) reduces the amplitude of the correction current command (Idqh*) in a specific operating region (E) set according to the target torque (T*). The specific operating region (E) is set to an operating region where the amount of phase change of the correction current command (Idqh*) relative to the change of the target torque (T*) is above a predetermined value.

[0111] According to this structure, in a specific operating region (E), the amplitude of the correction current command (Idqh*) decreases, and the correction torque used to suppress torque fluctuations also decreases. As a result, when the phase of the correction current command (Idqh*) deviates from the ideal phase, the possibility of the correction torque amplifying torque fluctuations can be reduced. The specific operating region (E) is set as the operating region where the change in phase of the correction current command (Idqh*) relative to the change in target torque (T*) is greater than a predetermined value; such a specific operating region (E) corresponds to a region where the phase deviation of the correction current command (Idqh*) is prone to increase. Therefore, by reducing the amplitude of the correction current command (Idqh*) in the specific operating region (E), the possibility of the correction torque amplifying torque fluctuations can be reduced, and torque fluctuations can be reduced in other operating regions through the correction torque. Thus, according to this structure, even in a rotating electric motor with torque variations, torque fluctuations can be appropriately reduced.

[0112] Furthermore, preferably, when a specific operating region (E) is set in this way, the time it takes for the phase of the correction current command (Idqh*) to be tracked to the phase corresponding to the changed target torque (T*) relative to the change of the target torque (T*) is taken as the convergence time. When the maximum amount of torque change that can be generated within the convergence time is generated, the specific operating region (E) is set according to the operating region that amplifies the torque fluctuation through the correction current command (Idqh*).

[0113] If a specific motion region (E) is set in this way, torque fluctuations can be appropriately reduced in motion regions that the control can track, and torque fluctuations can be amplified without correcting torque in motion regions that the control cannot track.

[0114] Furthermore, preferably, when the maximum torque change is generated within the said convergence time, and a specific action region (E) is set according to the action region that amplifies the torque vibration by the correction current command (Idqh*), the amplitude of the torque fluctuation of the reduced object is the same as the amplitude of the corrected torque based on the correction current command (Idqh*), and the specified value is ±60 degrees.

[0115] The maximum reduction effect can be achieved if the amplitude of the torque fluctuation reduction is the same as the amplitude of the corrected torque based on the corrected current command (Idqh*). However, with the optimal phase offset of the corrected torque, the reduction effect decreases as the offset increases, and the torque fluctuation is amplified when the phase offset exceeds 60 degrees. Therefore, if a predetermined value is set to 60 degrees, and the operating region where the phase change of the corrected current command (Idqh*) relative to the target torque (T*) is above a specified value is set as a specific operating region (E), then even if the target torque (T*) changes drastically, the torque fluctuation will not be amplified, thereby appropriately reducing the torque fluctuation.

[0116] In addition, preferably, the correction current command setting unit (6) sets the correction current command (Idqh*) when the rotational speed of the rotary motor (80) is a predetermined correction target speed.

[0117] By not superimposing the correction current command (Idqh*) in the entire operating region of the rotating electric motor 80, the efficiency reduction caused by the superposition of the correction current command (Idqh*) can be suppressed. Furthermore, in operating regions where torque fluctuations need to be suppressed, superimposing the correction current command (Idqh*) can appropriately suppress torque fluctuations.

[0118] Explanation of reference numerals in the attached figures:

[0119] 1: Torque control unit (basic current command setting unit), 2: Current control unit, 6: Correction current command setting unit, 7: Correction voltage command setting unit, 10: Rotating motor control device (rotating motor control system), 20: First current control unit, 21: Second current control unit, 80: Rotating motor, 84: Permanent magnet, 100: Rotating motor drive device (rotating motor control system), E: Specific operating area, Idq*: Basic current command, Idq**: Corrected current command (object current command), Idqh*: Correction current command, ST: Correction object speed, T*: Torque command (target torque), Vdq*: Voltage command, Vdqh*: Correction voltage command, θ: Magnetic pole position.

Claims

1. A rotary electric motor control system, wherein a permanent magnet type rotary electric motor driven by N-phase AC is used as the controlled object, where N is an arbitrary natural number, and current feedback control is performed in a dq-axis orthogonal vector coordinate system along the d-axis of the direction of the excitation flux generated by the permanent magnet and the q-axis orthogonal to the d-axis, wherein... The rotary motor control system has the following features: The basic current command setting unit sets a basic current command, which is the command value of the current flowing in the rotating motor, based on the target torque of the rotating motor. The correction current command setting unit sets a correction current command, which is a current command superimposed on the basic current command to reduce the torque fluctuation of the rotating motor, and is an alternating current with a frequency corresponding to the torque fluctuation of the target to be reduced. as well as The current control unit performs current feedback control based on the deviation between the target current command of the controlled object (obtained by adding the basic current command and the correction current command) and the current flowing through the rotating motor, i.e., the actual current. It also calculates the command value of the voltage applied to the rotating motor, i.e., the voltage command. The current control unit includes a first current control unit that controls the component of the basic current command, and a second current control unit that controls the component of the correction current command. The gain of the second current control unit is set to be lower than that of the first current control unit.

2. The rotary electric motor control system according to claim 1, wherein, The rotating motor control system has a correction voltage command setting unit, which sets a correction voltage command that is superimposed on the voltage command using feedforward control, in order to reduce torque fluctuations of the rotating motor.

3. The rotary electric motor control system according to claim 2, wherein, The correction voltage command setting unit sets the correction voltage command based on the target torque, the rotational speed of the rotary motor, and the magnetic pole position of the rotary motor.

4. The rotary electric motor control system according to claim 1, wherein, The correction current command is an alternating current whose phase differs according to the target torque of the rotating electric motor. The correction current command setting unit reduces the amplitude of the correction current command within a specific operating range set according to the target torque. The specific operating region is defined as the operating region in which the change in phase of the correction current command relative to the change in the target torque is above a predetermined value.

5. The rotary electric motor control system according to claim 4, wherein, The convergence time is defined as the time it takes for the phase of the correction current command to be tracked to the phase corresponding to the changed target torque, relative to the change in the target torque. In the case of the maximum torque change that can be generated within the said convergence time, the specific operating region is set according to the operating region that amplifies the torque fluctuation through the said correction current command.

6. The rotary electric motor control system according to claim 5, wherein, The amplitude of the torque fluctuation of the object being reduced is the same as the amplitude of the corrected torque based on the corrected current command, and the specified value is ±60 degrees.

7. The rotary electric motor control system according to any one of claims 1 to 6, wherein, The correction current command setting unit sets the correction current command when the rotational speed of the rotating motor is the predetermined correction target speed.

Images