Motor control device

Abstract

In a motor control device for controlling the rotation of an AC motor having two three-phase windings and a resolver (2) having two systems by finding voltage command values on dq axes, a first angle (θ1) found from the output of one system of the resolver (2) is used to perform dq conversion of the current of one three-phase winding, a second angle (θ2) found from the output of the other system of the resolver (2) is used to perform dq conversion of the current of the other three-phase winding, the d-axis current command values (id_ref) of both are made the same value, respective voltage command values on the dq axes are found, and the first angle (θ1) is used to convert to voltage command values of the voltage applied to one three-phase winding, and the second angle (θ2) is used to convert to voltage command values of the voltage applied to the other three-phase winding.

Description

TECHNICAL FIELD

[0001] The present application relates to a motor control device. BACKGROUND

[0002] As an angle detector that detects a rotation angle of a motor, a resolver is used most frequently. The resolver is known as a robust angle detector, but due to a requirement for a failure tolerance of a motor drive system, the resolver as an angle detector of the motor also needs to have redundancy. For example, in Patent Literature 1, a resolver is disclosed that has a wheel-shaped stator of resolver windings composed of an excitation winding and an output winding, and a rotor rotatably disposed inside the wheel-shaped stator, and is configured as a double system structure of a first system and a second system resolver winding provided on one wheel-shaped stator.

[0003] In addition, in Patent Literature 2, in paragraphs

[0043] to

[0044] and Figure 10 , it is disclosed that teeth of a stator are divided into 4 in a circumferential direction, and with respect to the 4, by being arranged in positions opposite to teeth that constitute one system as "first system first block B1", "second system first block B2", "first system second block B3", and "second system second block B4", it is possible to moderate an imbalance of magnetic flux when the stator is eccentric, and to improve angle detection accuracy.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT LITERATURE

[0006] Patent Literature 1: Japanese Patent Laid-Open No. 2000-18968

[0007] Patent Literature 2: International Publication No. WO2019 / 123592 SUMMARY

[0008] PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] As shown in Patent Literatures 1 and 2, in a case where teeth of a stator are divided into a plurality in a circumferential direction and the number of divisions is increased, depending on a direction of relative eccentricity of the stator and a rotor, an imbalance of magnetic resistance between the stator and the rotor cannot be eliminated, and an error in a detected angle occurs due to a primary component included in an output signal of the resolver, thereby causing a problem of torque pulsation caused by an angle error.

[0010] The present application discloses a technology for solving the above-mentioned problem, and aims to obtain a motor control device that can suppress pulsation caused by an angle error even if a resolver is eccentric.

[0011] TECHNICAL MEANS FOR SOLVING THE PROBLEMS

[0012] The motor control device disclosed in the present application controls the rotation of an alternating-current motor having a first three-phase winding and a second three-phase winding, includes a resolver in which a plurality of teeth provided in a stator are wound with a first field winding and a first output winding as a first system and a second field winding and a second output winding as a second system so that the number of the teeth wound with the windings of the first system is equal to the number of the teeth wound with the windings of the second system, and is characterized by including a first voltage application section that applies a three-phase alternating-current voltage to the first three-phase winding, a second voltage application section that applies a three-phase alternating-current voltage to the second three-phase winding, a first current detector that detects the current of each phase of the first three-phase winding, a second current detector that detects the current of each phase of the second three-phase winding, and a controller that outputs an alternating-current voltage applied to the first field winding, outputs an alternating-current voltage applied to the second field winding, and calculates and outputs to the first voltage application section a first voltage command value as a three-phase voltage command value in the first voltage application section, calculates and outputs to the second voltage application section a second voltage command value as a three-phase voltage command value in the second voltage application section, uses a first angle obtained from the output of the first system of the resolver to coordinate-convert three-phase current values detected by the first current detector into a first d-axis current value and a first q-axis current value, uses a second angle obtained from the output of the second system of the resolver to coordinate-convert three-phase current values detected by the second current detector into a second d-axis current value and a second q-axis current value, respectively, selects either a first d-axis current command value candidate calculated from a first current command value input as a current command value of the first three-phase winding or a second d-axis current command value candidate calculated from a second current command value input as a current command value of the second three-phase winding and sets it as a d-axis current command value, calculates a first d-axis voltage command value so that the d-axis current command value coincides with the first d-axis current value and calculates a first q-axis voltage command value so that a first q-axis current command value obtained from the first current command value coincides with the first q-axis current value, calculates the first voltage command value by coordinate-converting the first d-axis voltage command value and the first q-axis voltage command value using the first angle, calculates a second d-axis voltage command value so that the d-axis current command value coincides with the second d-axis current value and calculates a second q-axis voltage command value so that a second q-axis current command value obtained from the second current command value coincides with the second q-axis current value, and calculates the second voltage command value by coordinate-converting the second d-axis voltage command value and the second q-axis voltage command value using the second angle.

[0013] Further, a motor control device controls rotation of an alternating-current motor having a first three-phase winding and a second three-phase winding, and includes a resolver in which a first field winding and a first output winding as a first system and a second field winding and a second output winding as a second system are wound on a plurality of teeth provided in a stator so that the number of the teeth on which the windings of the first system are wound is equal to the number of the teeth on which the windings of the second system are wound, the motor control device being characterized by comprising: a first voltage application section that applies a three-phase alternating-current voltage to the first three-phase winding; a second voltage application section that applies a three-phase alternating-current voltage to the second three-phase winding; a first current detector that detects a current of each phase of the first three-phase winding; a second current detector that detects a current of each phase of the second three-phase winding; and a controller that outputs an alternating-current voltage applied to the first field winding, outputs an alternating-current voltage applied to the second field winding, and calculates and outputs, to the first voltage application section, a first voltage command value that is a three-phase voltage command value in the first voltage application section, calculates and outputs, to the second voltage application section, a second voltage command value that is a three-phase voltage command value in the second voltage application section, the controller using a correction angle set to an average value of a first angle obtained from an output of the first system of the resolver and a second angle obtained from an output of the second system, coordinate-converting a three-phase current value detected by the first current detector into a first d-axis current value and a first q-axis current value, coordinate-converting a three-phase current value detected by the second current detector into a second d-axis current value and a second q-axis current value, calculating a first d-axis voltage command value so that a first d-axis current command value obtained from a first current command value input as a current command value of the first three-phase winding coincides with the first d-axis current value, and calculating a first q-axis voltage command value so that a first q-axis current command value obtained from the first current command value coincides with the first q-axis current value, calculating the first voltage command value by coordinate-converting the first d-axis voltage command value and the first q-axis voltage command value using the correction angle, calculating a second d-axis voltage command value so that a second d-axis current command value obtained from a second current command value input as a current command value of the second three-phase winding coincides with the second d-axis current value, and calculating a second q-axis voltage command value so that a second q-axis current command value obtained from the second current command value coincides with the second q-axis current value, calculating the second voltage command value by coordinate-converting the second d-axis voltage command value and the second q-axis voltage command value using the correction angle.

[0014] Inventive Effects

[0015] According to the motor control device disclosed in the present application, a motor control device capable of suppressing pulsation caused by angle error even when the rotary transformer generates eccentricity can be obtained. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a block diagram showing the structure of the motor control device according to Embodiment 1.

[0017] Figure 2 is a diagram showing one example of the wiring of the winding of the alternating-current motor which is the control target of the motor control device disclosed in the present application.

[0018] Figure 3 is a conceptual diagram showing the structure of the rotary transformer used in the motor control device disclosed in the present application.

[0019] Figure 4 is a structural diagram of the rotary transformer used in the motor control device disclosed in the present application, obtained by viewing it from the direction perpendicular to the shaft.

[0020] Figure 5 is a diagram for explaining the operation of the rotary transformer used in the motor control device disclosed in the present application.

[0021] Figure 6 is a block diagram showing the structure of the controller of the motor control device according to Embodiment 1.

[0022] Figure 7 is a diagram showing the angle error of the rotary transformer for explaining the operation of the motor control device according to Embodiment 1.

[0023] Figure 8 is a block diagram showing the structure of the motor control device according to Embodiment 2.

[0024] Figure 9 is a block diagram showing the structure of the first controller of the motor control device according to Embodiment 2.

[0025] Figure 10 is a diagram showing the output signal of the rotary transformer for explaining the operation of the first controller of the motor control device according to Embodiment 2.

[0026] Figure 11 is a diagram showing the frequency analysis result of the output signal of the rotary transformer for explaining the operation of the first controller of the motor control device according to Embodiment 2.

[0027] Figure 12 is a block diagram showing the structure of the first removal processing section of the first controller of the motor control device according to Embodiment 2.

[0028] Figure 13 is a block diagram showing the structure of the second controller of the motor control device according to Embodiment 2.

[0029] Figure 14 is a graph showing the output signal of the resolver for explaining the operation of the second controller of the motor control device according to Embodiment 2.

[0030] Figure 15 is a graph showing the frequency analysis result of the output signal of the resolver for explaining the operation of the second controller of the motor control device according to Embodiment 2.

[0031] Figure 16 is a block diagram showing the structure of the second removal processing section of the second controller of the motor control device according to Embodiment 2.

[0032] Figure 17 is a block diagram showing the structure of the controller of the motor control device according to Embodiment 3.

[0033] Figure 18 is a block diagram showing the structure of the controller of the motor control device according to Embodiment 4.

[0034] Figure 19 is a block diagram showing the structure of the motor control device according to Embodiment 5.

[0035] Figure 20 is a block diagram showing the structure of the first controller of the motor control device according to Embodiment 5.

[0036] Figure 21 is a block diagram showing the structure of the second controller of the motor control device according to Embodiment 5.

[0037] Figure 22 is a block diagram showing the structure of the motor control device according to Embodiment 6.

[0038] Figure 23 is a block diagram showing the structure of the first controller of the motor control device according to Embodiment 6.

[0039] Figure 24 is a block diagram showing the structure of the second controller of the motor control device according to Embodiment 6.

[0040] Figure 25 is a block diagram showing one example of the specific structure of the controller of the motor control device according to the present disclosure. DETAILED DESCRIPTION

[0041] Embodiment 1.

[0042] Figure 1 is a block diagram of the motor control device according to Embodiment 1. The alternating-current motor 1 that is a control target is a permanent magnet synchronous rotating machine, a field winding synchronous rotating machine, an induction rotating machine, a synchronous reluctance motor, or the like, and has two three-phase windings, i.e., a first three-phase winding U1, V1, W1 and a second three-phase winding U2, V2, W2. The first three-phase winding U1-V1-W1 and the second three-phase winding U2-V2-W2 are wired, for example, as shown in Figure 2 Figure 2 In Figure 2 , as one example, a Y connection is shown, but a Δ connection can also be used. In addition, the first three-phase winding and the second three-phase winding can also have a phase difference.

[0043] Figure 3 is a conceptual diagram showing the structure of the resolver 2 used in the motor control device according to the present disclosure. As shown in Figure 3 , the resolver 2 has a first field winding 10A as a first system, two first output windings 111A, 112A, and a second field winding 10B as a second system, two second output windings 111B, 112B, as two systems.

[0044] Figure 4 is a structure diagram of the resolver 2 as viewed from a direction perpendicular to the axis. As shown in Figure 4 , the teeth TE1-TE12 are provided on the same stator 13, and the first field winding 10A, the two first output windings 111A, 112A, the second field winding 10B, and the two second output windings 111B, 112B are wound around the teeth TE1-TE12. At this time, the number of teeth around which the windings of the first system are wound is equal to the number of teeth around which the windings of the second system are wound. The rotor 14 is disposed radially inward of the stator 13. The rotor 14 includes a plurality of protruding portions that are uniformly disposed in the circumferential direction of the outer peripheral portion. The protruding height of the protruding portions toward the radial direction outside is formed in such a manner that the gap permeance between the stator 13 and the rotor 14 changes in a sinusoidal waveform in correspondence with rotation. Such a resolver 2 is called a variable reluctance (VR) type resolver. In Figure 4 , as an example of the resolver 2, a resolver provided with five protruding portions and having an axis multiple angle of five is illustrated. Thus, the resolver in which the rotor rotates five times in electrical angle when the rotor rotates one revolution is described as an example.

[0045] Figure 5An example of the operation of the resolver 2 assuming that there is no magnetic interference between the systems is shown. In a state where an alternating voltage VRA of a period TA is applied to the first exciting winding 10A, when the rotor rotates, the amplitude of the output signal VI A induced by the first output winding 111A and the amplitude of the output V2A induced by the first output winding 112A become sinusoidal (or cosinusoidal) according to the rotation angle (air gap permeance) at the electrical angle of the rotor. The first output winding 111A and the first output winding 112A are wound to the circumferential positions of the stator 13 in such a manner that the amplitudes of their alternating voltages differ by 90 degrees from each other in the electrical angle. Similarly, the second output winding 111B and the second output winding 112B are wound to the circumferential positions of the stator in such a manner that the amplitudes of their output signals VIB, V2B differ by 90 degrees from each other in the electrical angle.

[0046] The first exciting winding 10A wound to the plurality of teeth provided to the stator 13 is connected in series between the teeth, and the terminals of the first exciting winding 10A connected in series are connected to the controller 6 described later. Similarly, the terminals of the first output winding 111A connected in series between the teeth are connected to the controller 6 described later. The terminals of the other first output winding 112A connected in series between the teeth are connected to the controller 6 described later. The terminals of the second exciting winding 10B connected in series are connected to the controller 6 described later. Similarly, the terminals of the second output winding 111B connected in series between the teeth are connected to the controller 6 described later. The terminals of the other second output winding 112B connected in series between the teeth are connected to the controller 6 described later.

[0047] The direct current power supply 3a outputs a direct current voltage Vdcl to the first voltage application portion 4a. In addition, the direct current power supply 3b outputs a direct current voltage Vdc2 to the second voltage application portion 4b. These direct current power supplies can be any direct current power supplies that output a direct current voltage, such as a battery, a DC-DC converter, a diode rectifier, a PWM rectifier, and the like. Here, an example in which two direct current power supplies 3a and 3b are used is described, but one direct current power supply can be used, and the direct current voltage Vdc output from the one direct current power supply can be supplied to the first voltage application portion 4a and the second voltage application portion 4b.

[0048] The first voltage application section 4a and the second voltage application section 4b are each configured as an inverter that converts a direct current voltage into an alternating current voltage. The first voltage application section 4a performs PWM modulation by comparison with a PWM carrier having a frequency fc based on the first voltage command values Vu1_ref, Vv1_ref, Vw1_ref described later, thereby performing inversion conversion (conversion from direct current to alternating current) of the direct current voltage Vdcl input from the direct current power supply 3a, and applies alternating voltages Vu1, Vv1, Vw1 to the first three-phase winding U1, V1, W1 of the alternating current motor 1, respectively. As the switches Sup1 to Swn1 of the inverter configuring the first voltage application section 4a, semiconductor switches such as IGBT, bipolar transistor, MOS power transistor, and the like are used in reverse-parallel connection with diodes. The second voltage application section 4b performs PWM modulation by comparison with a PWM carrier having a frequency fc based on the second voltage command values Vu2_ref, Vv2_ref, Vw2_ref described later, thereby performing inversion conversion (conversion from direct current to alternating current) of the direct current voltage Vdc2 input from the direct current power supply 3b, and applies alternating voltages Vu2, Vv2, Vw2 to the second three-phase winding U2, V2, W2 of the alternating current motor 1, respectively. As the switches Sup2 to Swn2 of the inverter configuring the second voltage application section 4b, semiconductor switches such as IGBT, bipolar transistor, MOS power transistor, and the like are used in reverse-parallel connection with diodes.

[0049] The first current detector 5a detects each of the currents Iu1, Iv1, Iw1 flowing through the first three-phase winding U1, V1, W1, and outputs to the controller 6 described later. Similarly, the second current detector 5b detects each of the currents Iu2, Iv2, Iw2 flowing through the second three-phase winding U2, V2, W2, and outputs to the controller 6 described later. As these current detectors, a current detector of a known lower arm shunt method or bus 1 shunt method can be used.

[0050] The controller 6 inputs the first current command value I_targetl, the currents Iul, Ivl, Iwl detected by the first current detector 5a, and the output signals VlA of the first output winding 111A and V2A of the first output winding 112A of the resolver 2, and outputs the first voltage command values Vu1_ref, Vvl_ref, and Vwl_ref to the first voltage applying section 4a. The controller 6 also applies an alternating voltage VRA to the first exciting winding 10A of the resolver 2. Further, the controller 6 inputs the second current command value I_target2, the currents Iu2, Iv2, Iw2 detected by the second current detector 5b, and the output signals VIB of the second output winding 111B and V2B of the second output winding 112B of the resolver 2, and outputs the second voltage command values Vu2_ref, Vv2_ref, and Vw2_ref to the second voltage applying section 4b. The controller 6 also applies an alternating voltage VRB to the second exciting winding 10B of the resolver 2.

[0051] Next, the resolver 2 is excited by the controller 6. Figure 6 The configuration of the controller 6 will be described in detail. Figure 6 is a block diagram showing the operation of the controller 6. The first exciting section 601A applies an alternating voltage VRA of a sine wave of a period T to the first exciting winding 10A. Here, the first exciting section 601A can generate a rectangular wave signal of a period T that alternately outputs a binary voltage of an "H" level (e.g., 5 V) and an "L" level (e.g., 0 V) by a drive circuit, input the output to a low-pass filter circuit, and apply the output of the low-pass filter circuit as the alternating voltage VRA. Similarly, the second exciting section 601B applies an alternating voltage VRB of a sine wave of the same phase with the same period as the first exciting winding 10A to the second exciting winding 10B. Here, the second exciting section 601B can generate a rectangular wave signal of a period T that alternately outputs a binary voltage of an "H" level (e.g., 5 V) and an "L" level (e.g., 0 V) by a drive circuit, input the output to a low-pass filter circuit, and apply the output of the low-pass filter circuit as the alternating voltage VRB.

[0052] The first output signal detection section 602A periodically detects the output signals VlA and V2A of the two first output windings 111A and 112A of the first system of the resolver 2 at a predetermined detection timing (hereinafter also referred to as first detection timing) (see FIG. 6). Figure 5 The detected signals VlA_S and V2A_S are input to the first angle calculation section 604A, and the first angle θl is obtained from the operation of Equation (1).

[0053] θl = arctan(VlA_S / V2A_S) (1)

[0054] The second output signal detection section 602B periodically detects the output signals V1B, V2B of the two second output windings 111B, 112B of the second system of the resolver 2 at a predetermined detection timing (hereinafter also referred to as a second detection timing). The detected signals V1B_S, V2B_S are input to the second angle calculation section 604B, and the second angle θ2 is obtained from the operation of Equation (2).

[0055] θ2 = arctan(V1B_S / V2B_S) (2)

[0056] In the first angle correction section 605A, the average of the first angle θ1 and the second angle θ2 is operated, and output as a first correction angle θ1_m. Also, in the second angle correction section 605B, the average of the second angle θ2 and the first angle θ1 is operated, and output as a second correction angle θ2_m. That is, the first correction angle and the second correction angle are set to the same correction angle θ1_m = θ2_m = (θ1 + θ2) / 2.

[0057] The effects of the first angle correction section 605A and the second angle correction section 605B will be described later. The first current command value calculator 7a calculates a first q-axis current command value iq1_ref on the q-axis and a first d-axis current command value candidate id1_ref on the d-axis for energizing the first three-phase winding of the AC motor 1, based on the first current command value I_target1. The first current command value calculator 7a can substitute id1_ref = 0, iq1_ref = I_target1, can calculate using a known field weakening control, can calculate based on a known MTPA (Max Torque per Ampere) control, or can use other known methods of dq-axis current command calculation. Also, the second current command value calculator 7b calculates a second q-axis current command value iq2_ref on the q-axis and a second d-axis current command value candidate id2_ref on the d-axis for energizing the second three-phase winding of the AC motor 1, based on the second current command value I_target2. The second current command value calculator 7b can substitute id2_ref = 0, iq2_ref = I_target2, can calculate using a known field weakening control, can calculate based on a known MTPA (Max Torque per Ampere) control, or can use other known methods of dq-axis current command calculation.

[0058] The d-axis current selection section 606 inputs the output value of the first current command value calculator 7a, that is, the first d-axis current command value candidate idl_ref, and the output value of the second current command value calculator 7b, that is, the second d-axis current command value candidate id2_ref. In the d-axis current selection section 606, one of idl_ref and id2_ref is selected as the d-axis current command value id_ref and is output. The selection method can be set to select the one having a larger absolute value, or can be selected by another selection method.

[0059] The coordinate converter 8a performs coordinate conversion on the currents Iul, Ivl, Iwl flowing through the first three-phase winding detected by the first current detector 5a using the first correction angle θl_m output from the first angle correction section 605A, and obtains a current in a rotating two-axis (d-q axis), that is, a first d-axis current value idl and a first q-axis current value iq l. In addition, the coordinate converter 8b performs coordinate conversion on the currents Iu2, Iv2, Iw2 flowing through the second three-phase winding detected by the second current detector 5b using the second correction angle θ2_m output from the second angle correction section 605B, and obtains a current in a rotating two-axis (d-q axis), that is, a second d-axis current value id2 and a second q-axis current value iq2.

[0060] The subtracter 9a subtracts the d-axis current command value id_ref from the first d-axis current value idl and outputs a deviation err_dl. The subtracter 10a subtracts the first q-axis current command value iq l_ref from the first q-axis current value iq l and outputs a deviation err_q l. The subtracter 9b subtracts the d-axis current command value id_ref from the second d-axis current value id2 and outputs a deviation err_d2. The subtracter 10b subtracts the second q-axis current command value iq2_ref from the second q-axis current value iq2 and outputs a deviation err_q2.

[0061] The current controller 11a operates a first d-axis voltage command value vd1 on the rotating two-axes (d-q axes) using proportional integral control so that err_d1 obtained from the subtracter 9a coincides with zero, that is, a d-axis current command value id_ref coincides with a first d-axis current value id1. The current controller 12a operates a first q-axis voltage command value vq1 on the rotating two-axes (d-q axes) using proportional integral control so that err_q1 obtained from the subtracter 10a coincides with zero, that is, a first q-axis current command value iq1_ref coincides with a first q-axis current value iq1. The current controller 11b operates a second d-axis voltage command value vd2 on the rotating two-axes (d-q axes) using proportional integral control so that err_d2 obtained from the subtracter 9b coincides with zero, that is, a d-axis current command value id_ref coincides with a second d-axis current value id2. The current controller 12b operates a second q-axis voltage command value vq2 on the rotating two-axes (d-q axes) using proportional integral control so that err_q2 obtained from the subtracter 10b coincides with zero, that is, a second q-axis current command value iq2_ref coincides with a second q-axis current value iq2.

[0062] The coordinate converter 13a performs coordinate conversion on the first d-axis voltage command value vd1 and the first q-axis voltage command value vq1 on the rotating two-axes (d-q axes) using the first correction angle θ1_m, and operates three-phase voltage command values Vu1_ref, Vv1_ref, Vw1_ref output to the first voltage applying section 4a. The coordinate converter 13b performs coordinate conversion on the second d-axis voltage command value vd2 and the second q-axis voltage command value vq2 on the rotating two-axes (d-q axes) using the second correction angle θ2_m, and operates three-phase voltage command values Vu2_ref, Vv2_ref, Vw2_ref output to the second voltage applying section 4b.

[0063] Hereinafter, the reason why torque pulsation caused by angle error generated by eccentricity of the resolver can be suppressed will be explained using the present embodiment. Figure 7 A indicates that, in the case where there is no eccentricity, Figure 7B represents the result of frequency analysis (FFT) of each of the angle error of the first angle θ1, the second angle θ2, and the average of θ1 and θ2 in the case of eccentricity. The vertical axis is the angle error, and the horizontal axis is the mechanical angle order. In the case of eccentricity, the angle error of the first angle θ1 and the second angle θ2 becomes larger than in the case of no eccentricity at the mechanical angle 1st order, 4th order, 9th order, and 11th order. On the other hand, the angle error of the average of θ1 and θ2 is large at the 4th order and small at the 1st order, 9th order, and 11th order. It is considered that at the 1st order, 9th order, and 11th order where the error of the average of θ1 and θ2 is small, θ1 and θ2 have angle errors of the same magnitude and opposite signs, respectively, and thus the angle errors cancel each other out when the average of θ1 and θ2 is taken. In a resolver having two systems, when the number of teeth wound by each system is equal, a mechanical angle order component in which the angle error of the first angle θ1 and the angle error of the second angle θ2 are of the same magnitude and opposite signs is generated due to eccentricity of the resolver. Hereinafter, as described above, a structure for suppressing torque pulsation generated by the angle error of the order in which θ1 and θ2 have angle errors of the same magnitude and opposite signs is described.

[0064] When θ1 and θ2 have angle errors of the same magnitude and opposite signs, respectively, θ1 and θ2 are expressed by formula (3) and formula (4).

[0065] θ1 = θ + Δθ (3)

[0066] θ2 = θ - Δθ (4)

[0067] Here, let θ be the true angle, and Δθ be the angle error. The first angle correction unit 605A calculates the first corrected angle θ1_m using θ1 and θ2 according to formula (5).

[0068] θ1_m = 0.5 x (θ1 + θ2) (5)

[0069] By the calculation of formula (5), the angle error Δθ disappears, and θ1_m coincides with the value of the true angle θ. Also, the second angle correction unit 605B similarly calculates the second corrected angle θ2_m using θ2 and θ1 according to formula (6).

[0070] θ2_m = 0.5 x (θ2 + θ1) (6)

[0071] By the calculation of formula (6), the angle error Δθ disappears, and θ2_m coincides with the value of the true angle θ. As described above, by performing angle correction that cancels the angle error Δθ, torque pulsation caused by the angle error Δθ can be suppressed.

[0072] In the case where the angle error Δθ is a time-varying flux, if there is a difference between the first detection timing at which θ1 is detected and the second detection timing at which θ2 is detected, the angle error Δθ can not be canceled in the correction of the formula (5), the formula (6). This is because the angle error at the timing at which θ1 is detected and the angle error at the timing at which θ2 is detected differ in size. A method of suppressing the torque pulsation caused by the angle error in this case will be described. First, attention is paid to θ1. θ1 overlaps the true angle θ by the angle error + Δθ. In addition, the sign of Δθ itself is not positive or negative. Here, the axis on which dq conversion is performed using θ1 is set to be the dc1-qc1 axis. This means that the dc1-qc1 axis is angularly deviated from the true d-q axis by Δθ. At this time, the current id_c1 that flows in response to the command to the dc1 axis has a true q-axis component. Since the current having the true q-axis component contributes to torque, a q-axis current error Δiq1 is generated as the q-axis component of id_c1, id_c1 • sin(Δθ), thereby generating torque pulsation. In addition, the current iq_c1 that flows in response to the command to the qc1 axis also has an error in the true q-axis component, but the size thereof is iq_c1 • (1 - cos(Δθ)), and the influence thereof is almost negligible in the case where Δθ is small. In addition, the true d-axis component of iq_c1 contributes less to torque than the true q-axis component of id_c1, and thus is not considered here.

[0073] Next, attention is paid to θ2. θ2 overlaps the true angle θ by the angle error - Δθ. Here, the axis on which dq conversion is performed using θ2 is set to be the dc2-qc2 axis. This means that the dc2-qc2 axis is angularly deviated from the true d-q axis by - Δθ. At this time, the current id_c2 that flows in response to the command to the dc2 axis has a true q-axis component, the component of which becomes id_c2 • sin(- Δθ) and generates a q-axis current error Δiq2, thereby becoming a cause of torque pulsation. In addition, as in the case of the dc1-qc1 axis, the true q-axis component of the current iq_c2 that flows in response to the command to the qc2 axis has an influence that is almost negligible in the case where the angle error is small. In addition, the true d-axis component of iq_c2 contributes less to torque than the true q-axis component of id_c2, and thus is not considered here.

[0074] As described above, the q-axis current errors that contribute greatly to torque are the sum of Δiq1 and Δiq2. Here, when the sum of Δiq1 and Δiq2 is calculated, it is expressed by the formula (7).

[0075] Δiq1 + Δiq2 = (id_c1 - id_c2) sin(Δθ) (7)

[0076] In formula (7), it is known that the q-axis current error can be made zero by setting id cl and id c2 to the same value. Therefore, the torque ripple can be suppressed by setting id cl and id c2 to the same value. Therefore, by the d-axis current selection section 606, the d-axis current command value of the first three-phase winding and the d-axis current command value of the second three-phase winding are made equal, so that the torque ripple due to the angle error Δθ can be suppressed.

[0077] As explained above, in a case where the difference between the first detection timing and the second detection timing is small to be not problematic, the torque ripple due to eccentricity of the resolver can be reduced by the angle correction by the first angle correction section 605A and the second angle correction section 605B. On the other hand, in a case where the first detection timing and the second detection timing can have a difference, by setting the d-axis current command value of the first three-phase winding and the d-axis current command value of the second three-phase winding to the same value, the torque ripple due to eccentricity of the resolver can be reduced.

[0078] In addition, even in a case where the correction of the angle error by the first angle correction section 605A and the second angle correction section 605B is effective, of course, the d-axis current command value of the first three-phase winding and the d-axis current command value of the second three-phase winding can be made identical by the d-axis current selection section. Thereby, for performing control using the average value of θ1 and θ2 as the correction angle in the coordinate transformation and performing control using the same value as the d-axis current command value, by performing at least one of them, the torque ripple due to eccentricity of the resolver can be reduced.

[0079] Embodiment 2

[0080] Figure 8 is a block diagram of the motor control device according to Embodiment 2. The overlapping parts with Embodiment 1 are omitted from the explanation. Embodiment 2 is different from Embodiment 1 in that, as the controller, there are two controllers, the first controller 6a and the second controller 6b.

[0081] The first controller 6a inputs the first current command value I target1, the currents Iu1, IV1, Iw1 detected from the first current detector 5a, and the output signals V1A of the first output winding 111A and V2A of the first output winding 112A of the resolver 2, and outputs the first voltage command values Vu1_ref, Vv1_ref, and Vw1_ref to the first voltage application section 4a composed of an inverter. Further, the first periodic AC voltage VRA is applied to the first exciting winding 10A of the resolver 2.

[0082] The second controller 6b receives the first current command value I_target2, the currents Iu2, Iv2, and Iw2 detected by the second current detector 5b, and the output signals V1B and V2B of the second output winding 111B and second output winding 112B of the resolver 2. It then outputs the second voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref to the second voltage application unit 4b, which is composed of an inverter. Furthermore, an AC voltage VRB with a second period TB (different from the first period TA) is applied to the second excitation winding 10B of the resolver 2. Here, the relationship between the first period TA and the second period TB is assumed to be TB = 2 × TA.

[0083] Figure 9 This is a block diagram representing the operation of the first controller 6a. The first excitation unit 601A applies an AC voltage VRA (in this example, a sinusoidal AC voltage VRA) of the first cycle TA to the first excitation winding 10A. Specifically, the first excitation unit 601A can generate a rectangular wave signal of the first cycle TA that alternately outputs a binary voltage of "H" level (e.g., 5V) and "L" level (e.g., 0V) via a drive circuit, input this output to a low-pass filter circuit, and apply the output of the low-pass filter circuit as the AC voltage VRA.

[0084] The first output signal detection unit 602A periodically detects the output signals V1A and V2A of the two first output windings 111A and 112A at a preset detection time (hereinafter also referred to as the first detection time).

[0085] Here, we first explain the magnetic interference between the first output winding and the second output winding in the rotary transformer 2. The output signals V1A and V2A of the two first output windings 111A and 112A are respectively superimposed with second-cycle components V1A_TB and V2A_TB induced by the magnetic flux of the second cycle TB. The magnetic flux of the second cycle TB is obtained by the excitation of the second excitation winding 10B due to the magnetic interference between the systems. Figure 10 This indicates the output signal of the first output winding 111A. Figure 10 The top graph shows the output signal V1A of the first output winding 111A. The middle graph shows the component V1A_TA of the first cycle TA induced by the magnetic flux of the first excitation winding 10A included in the output signal V1A of the first output winding 111A. The bottom graph shows the component V1A_TB of the second cycle TB induced by the magnetic flux of the second excitation winding 10B included in the output signal V1A of the first output winding 111A. The output signal V1A of the first output winding 111A is obtained by summing the first cycle component V1A_TA and the second cycle component V1A_TB.

[0086] Here, Figure 11 A and Figure 11 B indicates a result of frequency analysis obtained by actually measuring the output signal V1A of the first output winding 111A. The actual measurement test conditions are TA = 50 us and TB = 100 us. Figure 11 A and Figure 11 In B, the horizontal axis indicates the frequency, and the vertical axis indicates the amplitude of the output signal. Figure 11 A indicates a case where the AC voltage of the period TB is applied to the second exciting winding 10B, and in the output signal V1A, V1A TB caused by the AC voltage of the period TB applied to the second exciting winding 10B is superimposed on V1A TA caused by the AC voltage of the period TA applied to the first exciting winding 10A as an interference voltage. On the other hand, Figure 11 B shows a case where the AC voltage of the period TB is not applied to the second exciting winding 10B, and the component V1A TB of the period TB contained in the output signal V1A is almost 0. The same applies to V2A, V2A TA, and V2A TB.

[0087] Therefore, when the angle is calculated from the signal containing V1A TB and V2A TB, a detection error occurs. Therefore, in order to suppress the detection error of the angle, it is necessary to remove the component V1A TB of the second period from the output signal V1A of the first output winding of the first system. Therefore, the first removal processing section 603A performs the second period component removal processing of removing (reducing) the component of the second period TB from the detected values V1A S and V2A S of the output signals of the two first output windings. Then, the first angle calculation section 604A calculates the first angle θ1 based on the detected values V1A F and V2A F of the output signals of the two first output windings after the second period component removal processing.

[0088] In the present embodiment, the second period component removal processing is performed based on the principle explained below. As Figure 10As shown in the graph below, in the second-period component V1A_TB of the output signal of the first output winding, the phase is reversed by adding an integer multiple of the second period TB to half the second period TB / 2 (for example, half the second period TB / 2), thus the sign of the value is reversed. Therefore, as a second-period component removal process, the first removal processing unit 603A is configured to add the detection values ​​V1A_S and V2A_S of the output signals of the two first output windings detected at the current detection timing to the detection values ​​V1A_Sold and V2A_Sold of the output signals of the two output windings of the first system detected at a detection timing that is a first removal processing interval ΔT1 earlier than the current detection timing. The first removal processing interval ΔT1 is set as shown in equation (8). Here, M is an integer greater than or equal to 0. In this embodiment, M is set to 0, and the first removal processing interval ΔT1 is set to half the second period TB / 2.

[0089] ΔT1=TB / 2+TB×M (8)

[0090] The first removal processing unit 603A, for example, is as follows: Figure 12 The configuration is as shown. The first removal processing unit 603A includes a first delay unit 6031A, which delays the detected value V1A_S of the output signal of the first output winding by only a first removal processing interval ΔT1 and outputs it. The detected value V1A_S of the output signal of the first output winding is added to the output V1A_Sold of the first delay unit 6031A to calculate the detected value V1A_F of the output signal of the first output winding after the second periodic component removal processing. Similarly, the first removal processing unit 603A includes a second delay unit 6032A, which delays the detected value V2A_S of the output signal of the first output winding by only a first removal processing interval ΔT1 and outputs it. The detected value V2A_S of the output signal of the first output winding is added to the output V2A_Sold of the second delay unit 6032A to calculate the detected value V2A_F of the output signal of the first output winding after the second periodic component removal processing.

[0091] Then, the first angle calculation unit 604A calculates the first angle θ1 based on the detection values ​​V1A_F and V2A_F of the output signals of the two first output windings 111A and 112A after addition. According to this structure, the components of the two second cycles, with their positive and negative signs reversed, are added together, and the components of the two second cycles cancel each other out. Therefore, the second cycle component is removed from the detection values ​​V1A_F and V2A_F of the output signals of the two first output windings after addition. Furthermore, the first angle θ1 can be calculated with good accuracy based on the detection values ​​after removing the second cycle component.

[0092] In the present embodiment, the first angle calculation section 604A calculates the first angle θ1 by calculating the arctangent (inverse tangent function) of the ratio of the detected value V1A_F of the output signal of the first output winding 111A after the second periodic component removal processing to the detected value V2A_F of the output signal of the first output winding 112A, as shown in the following expression (9).

[0093] θ1 = arctan(V1A_F / V2A_F) (9)

[0094] The first angle θ1 is communicated to the second controller 6b through CPU communication. The first angle θ1 and the second angle θ2, which is a signal communicated from the second controller 6b described later, are input to the first angle correction section 605A. In the first angle correction section 605A, the average of θ1 and θ2 is calculated, and the value is output as the first corrected angle θ1_m.

[0095] Next, the first current command value calculator 7a calculates the first q-axis current command value iq1_ref on the q-axis and the first d-axis current command value candidate id1_ref on the d-axis for energizing the first three-phase winding of the alternating-current motor 1 based on the first current command value I_target1. The first current command value calculator 7a can substitute id1_ref = 0 and iq1_ref = I_target1, can calculate using a known field-weakening control, can calculate based on a known MTPA (Max Torque per Ampere) control, or can use other known methods of dq-axis current command calculation. The first d-axis current command value candidate id1_ref for energizing the first three-phase winding is communicated to the second controller 6b through CPU communication.

[0096] The first d-axis current selection section 606A inputs the output value of the first current command value calculator 7a, that is, the first d-axis current command value candidate id1_ref, and a signal communicated from the second controller 6b described later, that is, the second d-axis current command value candidate id2_ref. In the first d-axis current selection section 606A, one of id1_ref and id2_ref is selected and output as the d-axis current command value id_ref. The selection method can be set to select the one with the larger absolute value or the one with the smaller absolute value.

[0097] The coordinate converter 8a performs coordinate conversion on the currents Iu1, Iv1, Iw1 flowing through the first three-phase winding detected by the first current detector 5a using the first corrected angle θ1_m output from the first angle correction section 605A, to obtain the first d-axis current value id1 and the first q-axis current value iq1 as currents on a rotating two-axis (d-q axis).

[0098] The subtracter 9a subtracts the d-axis current command value id_ref from the first d-axis current value idl, and outputs a deviation err_d1. The subtracter 10a subtracts the first q-axis current command value iq1_ref from the current iq1, and outputs a deviation err_q1. The current controller 11a operates the first d-axis voltage command value vd1 on the rotating two-axis (d-q axis) by proportional integral control so that the err_d1 obtained from the subtracter 9a coincides with zero. The current controller 12a operates the first q-axis voltage command value vq1 on the rotating two-axis (d-q axis) by proportional integral control so that the err_q1 obtained from the subtracter 10a coincides with zero. The coordinate converter 13a performs coordinate conversion on the first d-axis voltage command value vd1 and the first q-axis voltage command value vq1 on the rotating two-axis (d-q axis) using the first correction angle θ1_m, and obtains the first voltage command values Vu1_ref, Vv1_ref, Vw1_ref.

[0099] Next, the second controller 6b will be described. Figure 13 is a block diagram showing the operation of the second controller 6b. The second excitation section 601B applies an alternating-current voltage VRB of a second period TB of a sine wave to the second excitation winding 10B. Here, the second excitation section 601B can generate a second period TB of a rectangular wave signal that alternately outputs a binary voltage of "H" level (e.g., 5 V) and "L" level (e.g., 0 V), input the output to a low-pass filter circuit, and apply the output of the low-pass filter circuit as the alternating-current voltage VRB.

[0100] The second output signal detection section 602B periodically detects the output signals V1B, V2B of the two second output windings 111B, 112B at a detection timing (hereinafter also referred to as a second detection timing) set in advance.

[0101] As described above, the first output winding and the second output winding in the resolver 2 generate magnetic interference. Figure 14 The output signal V1B of the second output winding 111B is shown. The output signals V1B, V2B of the two second output windings 111B, 112B have superimposed thereon, respectively, first period components V1B_TA, V2B_TA induced by the magnetic flux of the first period TA due to the magnetic interference between the systems and excited by the first excitation winding 10A. Figure 14The upper graph shows the output signal V1B of the second output winding 111B, the middle graph shows the component V1B_TB of the second period TB contained in the output signal V1B of the second output winding 111B and induced by the magnetic flux of the second exciting winding 10B, and the lower graph shows the component V1B_TA of the first period TA contained in the output signal V1B of the second output winding 111B and induced by the magnetic flux of the first exciting winding 10A. The output signal V1B of the second output winding is a signal obtained by adding the component V1B_TB of the second period and the component V1B_TA of the first period.

[0102] Here, Figure 15 A and Figure 15 The frequency analysis result obtained by actually measuring the output signal V1B is shown in FIGS. 10A and 10B. The actual measurement test conditions are TA = 50 us and TB = 100 us. Figure 15 A and Figure 15 In FIGS. 10A and 10B, the horizontal axis represents frequency, and the vertical axis represents the amplitude of the output signal. Figure 15 A shows a case where the AC voltage of the period TA is applied to the first exciting winding, and V1B_TA due to the AC voltage of the period TA applied to the first exciting winding is superimposed as an interference voltage into the output signal V1B in addition to V1B_TB due to the AC voltage of the period TB applied to the second exciting winding. On the other hand, Figure 15 B shows a case where the AC voltage of the period TA is not applied to the first exciting winding, and the component V1B_TA of the period TA contained in the output signal V1B is almost 0. The same applies to V2B, V2B_TB, and V2B_TA.

[0103] Therefore, when the angle is calculated by the signal containing V1B_TA and V2B_TA, a detection error occurs. Therefore, in order to suppress the detection error of the angle, it is necessary to remove the component V1B_TA of the first period from the output signal V1B of the second output winding. Therefore, the second removal processing section 603B performs the first period component removal processing of removing (reducing) the component of the first period from the detected values V1B_S and V2B_S of the output signals of the two second output windings 111B and 112B. Then, the second angle calculation section 604B calculates the second angle θ2 based on the detected values V1B_F and V2B_F of the output signals of the two second output windings 111B and 112B after the first period component removal processing.

[0104] In the present embodiment, the first period component removal processing is performed based on the principle explained below. As Figure 14As shown in the graph below, the first-cycle component V1B_TA of the output signal of the second output winding becomes equal in cycles that are integer multiples of the first cycle TA (e.g., the first cycle TA). Therefore, as a first-cycle component removal process, the second removal processing unit 603B is configured to perform a subtraction process, which calculates the difference between the detection values ​​V1B_S and V2B_S of the output signals of the two second output windings 111B and 112B detected at the current detection timing and the detection values ​​V1B_Sold and V2B_Sold of the output signals of the two second output windings 111B and 112B detected at a detection timing that is an interval ΔT2 earlier than the current detection timing. As shown in equation (10), the second system removal processing interval ΔT2 is set to an integer multiple of the first cycle TA. Here, P is an integer greater than or equal to 1. In this embodiment, P is set to 1, and the second system removal processing interval ΔT2 is set to the first cycle TA.

[0105] ΔT2=TA×P (10)

[0106] The second removal processing unit 603B is, for example, as follows: Figure 16 The second removal processing unit 603B is configured as shown. It includes a first delay unit 6031B, which delays the detected value V1B_S of the output signal of the second output winding by only a second removal processing interval ΔT2 and outputs it. It subtracts the output V1B_S of the first delay unit 6031B from the detected value V1B_S of the output signal of the second output winding 111B to calculate the detected value V1B_F of the output signal of the second output winding 111B after the first periodic component removal processing. Similarly, the second removal processing unit 603B includes a second delay unit 6032B, which delays the detected value V2B_S of the output signal of the second output winding 112B by only a second removal processing interval ΔT2 and outputs it. It subtracts the output V2B_S of the second delay unit 6032B from the detected value V2B_S of the output signal of the second output winding 112B to calculate the detected value V2B_F of the output signal of the second output winding 112B after the first periodic component removal processing.

[0107] Then, the second angle calculation unit 604B is configured to calculate the second angle θ2 based on the detection values ​​V1B_F and V2B_F of the output signals of the second output windings 111B and 112B after subtraction processing. According to this structure, the components of the two first cycles that become equal in value every second removal processing interval ΔT2 are subtracted and canceled out. Therefore, the first cycle component is removed from the detection values ​​V1B_F and V2B_F of the output signals of the two second output windings 111B and 112B after subtraction processing. Furthermore, the second angle θ2 can be calculated with good accuracy based on the detection values ​​after removing the first cycle component.

[0108] The second angle calculation section 604B calculates the second angle θ2 by calculating the arctangent (inverse tangent function) of the ratio of the detected value V1B_F of the output signal of the second output winding 111B after the first periodic component removal processing to the detected value V2B_F of the output signal of the second output winding 112B, as shown in the following expression (11).

[0109] θ2 = arctan(V1B_F / V2B_F) (11)

[0110] The second angle θ2 is communicated to the first controller 6a through CPU communication. The second angle θ2 and the first angle θ1, which is a signal communicated from the first controller 6a, are input to the second angle correction section 605B. In the second angle correction section 605B, the average value of θ2 and θ1 is calculated, and the value is output as the second corrected angle θ2_m.

[0111] Next, the second current command value calculator 7b calculates the second q-axis current command value iq2_ref on the q-axis and the second d-axis current command value candidate id2_ref on the d-axis for energizing the second three-phase winding of the alternating-current motor 1, based on the second current command value I_target2. The second current command value calculator 7b can substitute id2_ref = 0 and iq2_ref = I_target2, can calculate using a known field-weakening control, can calculate based on a known MTPA (Max Torque per Ampere) control, or can use other known methods of dq-axis current command calculation.

[0112] The second d-axis current command value candidate id2_ref for energizing the second three-phase winding is communicated to the first controller through CPU communication. The second d-axis current selection section 606B inputs the output value of the second current command value calculator 7b, that is, the second d-axis current command value candidate id2_ref, and a signal communicated from the first controller 6a, that is, the first d-axis current command value candidate id1_ref. In the second d-axis current selection section 606B, one of id2_ref and id1_ref is selected as the d-axis current command value id_ref and is output. The selection is the same selection as that of the first d-axis current selection section 606A.

[0113] The coordinate converter 8b uses the second corrected angle θ2_m output from the second angle correction section 605B to perform coordinate conversion on currents Iu2, Iv2, Iw2 flowing through the second three-phase winding detected by the second current detector 5b, and obtains second d-axis current value id2 and second q-axis current value iq2 in the rotating two-axis (d-q axis). The subtracter 9b subtracts the d-axis current command value id_ref from the second d-axis current value id2, and outputs a deviation err_d2. The subtracter 10b subtracts the second q-axis current command value iq2_ref from the second q-axis current value iq2, and outputs a deviation err_q2. The current controller 11b performs proportional integral control operation on the second d-axis voltage command value vd2 in the rotating two-axis (d-q axis) so that err_d2 obtained from the subtracter 9b coincides with zero. The current controller 12b performs proportional integral control operation on the second q-axis voltage command value vq2 in the rotating two-axis (d-q axis) so that err_q2 obtained from the subtracter 10b coincides with zero. The coordinate converter 13b performs coordinate conversion on the second d-axis voltage command value vd2 and the second q-axis voltage command value vq2 in the rotating two-axis (d-q axis) using the second corrected angle θ2_m, and obtains second voltage command values Vu2_ref, Vv2_ref, Vw2_ref.

[0114] The reason why torque pulsation due to angle error caused by eccentricity of the rotary transformer is generated is the same as in Embodiment 1. The reason why torque pulsation can be suppressed is also the same as in Embodiment 1. In the structure of Embodiment 2, the first angle θ1, the second angle θ2, the first d-axis current command value candidate id1_ref, and the second d-axis current command value candidate id2_ref are communicated with each other between the first controller 6a and the second controller 6b. Here, the error Δθ included in the first angle θ1 and the second angle θ2, respectively, is an alternating current. Therefore, for example, if a communication delay occurs when the second angle θ2 communicates with the first controller 6a, a phase delay occurs in Δθ included in the second angle θ2. Therefore, even if the average of the first angle θ1 and the second angle θ2 transmitted from the second controller 6b in a communication manner is taken, Δθ does not cancel out, and thus it can not be possible to suppress angle error. The same applies in the case where the first angle θ1 is communicated from the first controller 6a to the second controller 6b. Even in this case, since the first d-axis current command value candidate id1_ref, the second d-axis current command value candidate id2_ref, and the d-axis current command value id_ref are direct currents, the influence of the communication delay is small, and thus torque pulsation due to angle error can be appropriately suppressed.

[0115] Embodiment 3.

[0116] Figure 17is a block diagram showing the structure of the controller 6 of the motor control device according to Embodiment 3. In the structure having only one controller 6 as in Embodiment 1, in the case where the detection timings of both the first output signal detection section 602A and the second output signal detection section 602B are the same, the d-axis current command value in the first three-phase winding and the d-axis current command value in the second three-phase winding can also be made different. In this case, as shown in Figure 17 the d-axis current selection section 606 explained in Embodiment 1 can also not be provided. Then, instead of the d-axis current command value id_ref explained in Embodiment 1, the first d-axis current command value candidate idl_ref output from the first current command value calculator 7a is directly used as the d-axis current command value in the first three-phase winding, that is, the first d-axis current command value, the first d-axis voltage command value vd1 is calculated, and the second d-axis current command value candidate id2_ref output from the second current command value calculator 7b is directly used as the d-axis current command value in the second three-phase winding, that is, the second d-axis current command value, the second d-axis voltage command value vd2 is calculated.

[0117] Even in this case, as explained in Embodiment 1, by performing angle correction on the first angle correction section 605A and the second angle correction section 605B using the equations (5) and (6), it is possible to eliminate the influence of eccentricity of the resolver.

[0118] Embodiment 4.

[0119] Figure 18 is a block diagram showing the structure of the controller 6 of the motor control device according to Embodiment 4. In the structure having only one controller as in Embodiment 1, in the case where the detection timings of the first output signal detection section 602A and the second output signal detection section 602B are known to be different in advance, the angle correction shown in the equations (5) and (6) can not be performed. In this case, as shown in Figure 18 the output signal of the first angle calculation section 604A, that is, the first angle θ1 is input to the coordinate converter 8a, 13a, and the output signal of the second angle calculation section 604B, that is, the second angle θ2 is input to the coordinate converter 8b, 13b to perform coordinate conversion. In this case, since the d-axis current command value in the first three-phase winding and the d-axis current command value in the second three-phase winding are made the same d-axis current command value id_ref by the d-axis current selection section 606, id_cl and id_c2 in the equation (7) explained in Embodiment 1 are set to the same value, and it is possible to eliminate the influence of eccentricity of the resolver.

[0120] Embodiment 5.

[0121] Figure 19is a block diagram of the motor control device according to Embodiment 5. As in Embodiment 2, the controller is constituted by two controllers, the first controller 6a and the second controller 6b, and in a case where the detection timings of the first output signal detection section 602A and the second output signal detection section 602B are the same and the communication delay between the two controllers is small, the d-axis current command value in the first three-phase winding and the d-axis current command value in the second three-phase winding can not be made to coincide. In this case, as shown in Figure 19 , the first d-axis current command value candidate idl_ref and the second d-axis current command value candidate id2_ref can not be communicated between the first controller 6a and the second controller 6b. That is, the first controller 6a as shown in Figure 20 , the second controller 6b as shown in Figure 21 , the first d-axis current selection section 606A and the second d-axis current selection section 606B in Embodiment 2 can not be provided. In the first controller 6a, the first d-axis current command value candidate idl_ref output from the first current command value calculator 7a is directly used as the d-axis current command value in the first three-phase winding, that is, the first d-axis current command value, to calculate the first d-axis voltage command value vdl. Also, in the second controller 6b, the second d-axis current command value candidate id2_ref output from the second current command value calculator 7b is directly used as the d-axis current command value in the second three-phase winding, that is, the second d-axis current command value, to calculate the second d-axis voltage command value vd2.

[0122] Even in this case, by performing angle correction in the first angle correction section 605A and the second angle correction section 605B, respectively, it is possible to eliminate the influence of eccentricity of the resolver.

[0123] Embodiment 6.

[0124] Figure 22 is a block diagram of the motor control device according to Embodiment 6. As in Embodiment 2, the controller is constituted by two controllers, the first controller 6a and the second controller 6b, and in a case where it is known in advance that the detection timings of the first output signal detection section 602A and the second output signal detection section 602B are different or the communication delay between the two controllers cannot be ignored, it is not necessarily required to perform angle correction shown by the equations (5) and (6). In this case, as shown in Figure 22 , the data of the first angle θl and the second angle θ2 can not be communicated between the first controller 6a and the second controller 6b. As shown in Figure 23 the first controller 6a and Figure 24As with the configuration of the second controller 6b shown, the first angle correction section 605A and the second angle correction section 605B provided in Embodiment 2 are omitted, the first angle θ1 as the output signal of the first angle calculation section 604A is input to the coordinate converters 8a, 13a, the second angle θ2 as the output signal of the second angle calculation section 604B is input to the coordinate converters 8b, 13b, and the coordinate conversion is performed.

[0125] Even in this case, the data of the first d-axis current command value candidate idl_ref and the second d-axis current command value candidate id2_ref are communicated between the first controller 6a and the second controller 6b, and either one of idl_ref and id2_ref is selected in the first d-axis current selection section 606A and the second d-axis current selection section 606B, and the d-axis current command value id_ref is set to both, whereby the influence of eccentricity of the resolver can be eliminated

[0126] For the controller 6, the first controller 6a, the second controller 6b in each of the above embodiments,

[0127] Specifically, as Figure 25As shown, an arithmetic processing device 101 such as a CPU (Central Processing Unit), a storage device 102 that exchanges data with the arithmetic processing device 101, an input / output interface 103 that inputs and outputs signals between the arithmetic processing device 101 and the outside, and the like are included. As the arithmetic processing device 101, an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), various signal processing circuits, and the like can be provided. In Embodiment 2 and the like, the controller 6a and the controller 6b can also be constituted by one arithmetic processing device 101. As the storage device 102, a RAM (Random Access Memory) configured to be able to read and write data from and to the arithmetic processing device 101, a ROM (Read Only Memory) configured to be able to read data from the arithmetic processing device 101, and the like can be provided. The input / output interface 103 is constituted by, for example, an A / D converter that inputs signals output from the resolver 2 and the current detector to the arithmetic processing device 101, a D / A converter that inputs a voltage command signal from the arithmetic processing device 101 to the first voltage applying section 4a and the second voltage applying section 4b, and the like.

[0128] Although various exemplary embodiments and examples are described in the present application, various features, modes, and functions described in one or more embodiments are not limited to the application of the specific embodiments, and can be applied to the embodiments individually or in various combinations. Therefore, it can be considered that numerous modifications not exemplified are also included in the technical scope disclosed in the present application specification. For example, cases in which at least one constituent element is modified, added, or omitted, and cases in which at least one constituent element is extracted and combined with the constituent elements of other embodiments are included.

[0129] Label Explanation

[0130] 1 AC motor

[0131] 2 Resolver

[0132] 4a First voltage applying section

[0133] 4b Second voltage applying section

[0134] 5a First current detector

[0135] 5b second current detector

[0136] 6 controller

[0137] 6a first controller

[0138] 6b second controller

[0139] 10A first field winding

[0140] 111A, 112A first output winding

[0141] 10B second field winding

[0142] 111B, 112B second output winding

[0143] U1, V1, W1 first three-phase winding

[0144] U2, V2, W2 second three-phase winding

Claims

1. An electric motor control device that controls rotation of an alternating-current electric motor having a first three-phase winding and a second three-phase winding, including a resolver in which a plurality of teeth provided in a stator are wound with a first field winding and a first output winding as a first system and a second field winding and a second output winding as a second system so that the number of the teeth wound with the windings of the first system is equal to the number of the teeth wound with the windings of the second system, the electric motor control device being characterized by comprising: a first voltage application section that applies a three-phase alternating-current voltage to the first three-phase winding; a second voltage application section that applies a three-phase alternating-current voltage to the second three-phase winding; a first current detector that detects a current of each phase of the first three-phase winding; a second current detector that detects a current of each phase of the second three-phase winding; and a controller that outputs an alternating-current voltage applied to the first field winding, outputs an alternating-current voltage applied to the second field winding, and calculates a first voltage command value as a three-phase voltage command value in the first voltage application section and outputs the first voltage command value to the first voltage application section, calculates a second voltage command value as a three-phase voltage command value in the second voltage application section and outputs the second voltage command value to the second voltage application section, the controller, using a first angle obtained from an output of the first system of the resolver, coordinate-converts three-phase current values detected by the first current detector into a first d-axis current value and a first q-axis current value, using a second angle obtained from an output of the second system of the resolver, coordinate-converts three-phase current values detected by the second current detector into a second d-axis current value and a second q-axis current value, respectively, selects either a first d-axis current command value candidate calculated from a first current command value input as a current command value of the first three-phase winding or a second d-axis current command value candidate calculated from a second current command value input as a current command value of the second three-phase winding and sets the selected one as a d-axis current command value, calculates a first d-axis voltage command value so that the d-axis current command value coincides with the first d-axis current value and calculates a first q-axis voltage command value so that a first q-axis current command value obtained from the first current command value coincides with the first q-axis current value, calculates the first voltage command value by coordinate-converting the first d-axis voltage command value and the first q-axis voltage command value using the first angle, calculates a second d-axis voltage command value so that the d-axis current command value coincides with the second d-axis current value and calculates a second q-axis voltage command value so that a second q-axis current command value obtained from the second current command value coincides with the second q-axis current value, and calculates the second voltage command value by coordinate-converting the second d-axis voltage command value and the second q-axis voltage command value using the second angle. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ 2. The motor control device according to claim 1, wherein the controller applies the same phase AC voltage to the first field winding and the second field winding with the same period.

3. The motor control device according to claim 1, wherein the controller applies a first period AC voltage to the first field winding and a second period AC voltage different from the first period to the second field winding, and the first angle is calculated by removing a component of the second period from an output signal of the first output winding, and the second angle is calculated by removing a component of the first period from an output signal of the second output winding.

4. The motor control device according to any one of claims 1 to 3, wherein the controller has a first controller that inputs the output of the first current detector and the first current command value and outputs the first voltage command value, and a second controller that inputs the output of the second current detector and the second current command value and outputs the second voltage command value.

5. A motor control device that controls rotation of an alternating-current motor having a first three-phase winding and a second three-phase winding, including a resolver in which a plurality of teeth provided in a stator are wound with a first field winding and a first output winding as a first system and a second field winding and a second output winding as a second system so that the number of the teeth wound with the windings of the first system is equal to the number of the teeth wound with the windings of the second system, the motor control device characterized by comprising: a first voltage application section that applies a three-phase AC voltage to the first three-phase winding; a second voltage application section that applies a three-phase AC voltage to the second three-phase winding; a first current detector that detects a current of each phase of the first three-phase winding; a second current detector that detects a current of each phase of the second three-phase winding; and a controller that outputs an AC voltage applied to the first field winding, outputs an AC voltage applied to the second field winding, and calculates a first voltage command value as a three-phase voltage command value in the first voltage application section and outputs to the first voltage application section, and calculates a second voltage command value as a three-phase voltage command value in the second voltage application section and outputs to the second voltage application section, the controller, using a correction angle set to an average of a first angle obtained from an output of the first system of the resolver and a second angle obtained from an output of the second system, coordinate-converts a three-phase current value detected by the first current detector into a first d-axis current value and a first q-axis current value, and coordinate-converts a three-phase current value detected by the second current detector into a second d-axis current value and a second q-axis current value. ​ ​ ​ ​ ​ ​ ​ ​ the first d-axis voltage command value is calculated so that a first d-axis current command value obtained from a first current command value input as a current command value for the first three-phase winding agrees with the first d-axis current value, and the first q-axis voltage command value is calculated so that a first q-axis current command value obtained from the first current command value agrees with the first q-axis current value, the first voltage command value is calculated by coordinate conversion of the first d-axis voltage command value and the first q-axis voltage command value using the correction angle, the second d-axis voltage command value is calculated so that a second d-axis current command value obtained from a second current command value input as a current command value for the second three-phase winding agrees with the second d-axis current value, and the second q-axis voltage command value is calculated so that a second q-axis current command value obtained from the second current command value agrees with the second q-axis current value, the second voltage command value is calculated by coordinate conversion of the second d-axis voltage command value and the second q-axis voltage command value using the correction angle.

6. The motor control device according to claim 5, wherein the controller applies an alternating-current voltage of the same phase to the first field winding and the second field winding with the same period.

7. The motor control device according to claim 5, wherein the controller applies an alternating-current voltage of a first period to the first field winding, applies an alternating-current voltage of a second period different from the first period to the second field winding, and the first angle is calculated by removing a component of the second period from an output signal of the first output winding, and the second angle is calculated by removing a component of the first period from an output signal of the second output winding.

8. The motor control device according to any one of claims 5 to 7, wherein the controller has a first controller that inputs an output of the first current detector and the first current command value, and outputs the first voltage command value, and a second controller that inputs an output of the second current detector and the second current command value, and outputs the second voltage command value.

9. A motor control device that controls rotation of an alternating-current motor having a first three-phase winding and a second three-phase winding, the motor control device including a resolver in which a first field winding and a first output winding as a first system and a second field winding and a second output winding as a second system are wound around a plurality of teeth provided in a stator so that the number of the teeth around which the windings of the first system are wound is equal to the number of the teeth around which the windings of the second system are wound, the motor control device characterized by comprising: a first voltage application section that outputs a three-phase alternating-current voltage applied to the first three-phase winding; a second voltage application section that outputs a three-phase alternating-current voltage applied to the second three-phase winding; a first current detector that detects a current of each phase of the first three-phase winding; a second current detector that detects a current of each phase of the second three-phase winding; a first angle calculation section that calculates a first angle based on an output of the first current detector and an output of the first voltage application section; a second angle calculation section that calculates a second angle based on an output of the second current detector and an output of the second voltage application section; a first voltage command calculation section that calculates a first voltage command value based on an output of the first angle calculation section; a second voltage command calculation section that calculates a second voltage command value based on an output of the second angle calculation section; a first field winding application section that applies an alternating-current voltage to the first field winding based on an output of the first voltage command calculation section; a second field winding application section that applies an alternating-current voltage to the second field winding based on an output of the second voltage command calculation section; a first output winding application section that applies an alternating-current voltage to the first output winding based on an output of the first angle calculation section; and a second output winding application section that applies an alternating-current voltage to the second output winding based on an output of the second angle calculation section. ​ ​ ​ a second current detector that detects currents of respective phases of the second three-phase winding; and a controller that outputs an alternating-current voltage applied to the first field winding, outputs an alternating-current voltage applied to the second field winding, and calculates a first voltage command value as a three-phase voltage command value in the first voltage application section and outputs to the first voltage application section, calculates a second voltage command value as a three-phase voltage command value in the second voltage application section and outputs to the second voltage application section, the controller, coordinates the three-phase current values detected by the first current detector into a first d-axis current value and a first q-axis current value using a correction angle set to an average value of a first angle obtained from an output of the first system of the resolver and a second angle obtained from an output of the second system, coordinates the three-phase current values detected by the second current detector into a second d-axis current value and a second q-axis current value, selects either a first d-axis current command value candidate obtained from a first current command value input as a current command value of the first three-phase winding or a second d-axis current command value candidate obtained from a second current command value input as a current command value of the second three-phase winding and sets as a d-axis current command value, calculates a first d-axis voltage command value to make the d-axis current command value coincide with the first d-axis current value and calculates a first q-axis voltage command value to make a first q-axis current command value obtained from the first current command value coincide with the first q-axis current value, calculates the first voltage command value by coordinate-converting the first d-axis voltage command value and the first q-axis voltage command value using the correction angle, calculates a second d-axis voltage command value to make the d-axis current command value coincide with the second d-axis current value and calculates a second q-axis voltage command value to make a second q-axis current command value obtained from the second current command value coincide with the second q-axis current value, calculates the second voltage command value by coordinate-converting the second d-axis voltage command value and the second q-axis voltage command value using the correction angle.

10. The motor control device according to claim 9, wherein the controller applies alternating-current voltages of the same phase to the first field winding and the second field winding with the same period.

11. The motor control device according to claim 9, wherein the controller applies an alternating-current voltage of a first period to the first field winding and applies an alternating-current voltage of a second period different from the first period to the second field winding, and the first angle is calculated by removing a component of the second period from an output signal of the first output winding, and the second angle is calculated by removing a component of the first period from an output signal of the second output winding.

12. The motor control device according to any one of claims 9 to 11, wherein The controller has a first controller which inputs the output of the first current detector and the first current command value and outputs the first voltage command value, and a second controller which inputs the output of the second current detector and the second current command value and outputs the second voltage command value.

Images