WO2019150984A1 - Control device for three-phase synchronous electric motor - Google Patents

Abstract

Provided is a control device for a three-phase synchronous electric motor that makes it possible to improve accuracy of detection of a rotor position even if there is offset included in a neutral point potential of the three-phase synchronous electric motor. A control device of a three-phase synchronous electric motor is provided with: a three-phase synchronous electric motor (4) provided with a first three-phase winding (41) and a second three-phase winding (42); a first inverter (31) connected to the first three-phase winding; a second inverter (32) connected to the second three-phase winding; a first control unit (61) for estimating a rotor position of the three-phase synchronous electric motor from a neutral point potential of the first three-phase winding and controlling the first inverter; and a second control unit (62) for estimating a rotor position of the three-phase synchronous electric motor from a neutral point potential of the second three-phase winding and controlling the second inverter. The first control unit corrects the neutral point potential of the first three-phase winding on the basis of a direct current voltage of the first inverter.

Description

Control device for three-phase synchronous motor

The present invention relates to a control device for a three-phase synchronous motor that controls the three-phase synchronous motor based on the position of the rotor.

Small and highly efficient three-phase synchronous motors (permanent magnet synchronous motors) are widely used in various fields such as industry, home appliances, and automobiles. In particular, in the field of automobile equipment such as an electric power steering device, a permanent magnet synchronous motor excellent in miniaturization and high efficiency is frequently used.

In a permanent magnet synchronous motor, in general, the rotational position of a rotor provided with a magnet is detected by a magnetic detection element such as a Hall IC, and based on the detection result, an armature coil on the stator side is sequentially excited to rotate the rotor. It is rotating. In addition, by using a resolver, encoder, GMR sensor (GMR: Giantto Magneto-Resistivity effect), which is a precise rotational position detector, it is possible to drive with sinusoidal current and reduce vibration and noise such as torque ripple. I am trying. In recent years, rotational position sensorless control that performs motor speed and torque control without providing this rotational position sensor has become widespread.

実 用 Practical use of rotational position sensorless control can reduce the cost of the position sensor (the cost of the sensor itself, the cost of the sensor wiring, etc.) and reduce the size of the device. Further, since the sensor is not required, there is an advantage that the motor can be controlled in a poor environment for the sensor.

Currently, the rotational position sensorless control of a permanent magnet synchronous motor directly detects an induced voltage (speed electromotive voltage) generated by the rotation of a rotor equipped with a magnet, and drives the permanent magnet synchronous motor as rotor position information. And a position estimation method for estimating and calculating the rotor position from a mathematical model of the target motor.

There are many problems with these rotational position sensorless controls. In general, a problem often described is a position detection method when the rotational speed of the motor is low. Most rotational position sensorless controls that are currently in practical use are based on an induced voltage (speed electromotive force) generated by the rotation of a permanent magnet synchronous motor. Therefore, in a stop where the induced voltage is small and in a low speed range, the sensitivity is lowered, and the position information may be buried in noise. As a solution to this problem, techniques described in Patent Documents 1 to 4 are known.

In the technique described in Patent Document 1, a high frequency current is applied to a permanent magnet synchronous motor, and a rotor position is detected from a current harmonic generated at that time and a mathematical model of the permanent magnet synchronous motor. In this technique, position detection is possible by using current harmonics generated by the saliency of the rotor of the permanent magnet synchronous motor.

In the technique described in Patent Document 2, an electromotive voltage (speed) generated in a non-conduction phase is based on a 120-degree conduction method in which two phases are selected and energized among three-phase stator windings of a permanent magnet synchronous motor. The position of the rotor is detected based on an electromotive voltage due to an inductance imbalance rather than an electromotive voltage associated with. In this technique, since an electromotive voltage generated according to a position is used, position information can be acquired even in a complete stop state.

In the technique described in Patent Document 3, “neutral point potential” which is the potential of the connection point of the three-phase stator winding is detected to obtain position information. At that time, by detecting the neutral point potential in synchronization with the PWM (pulse width modulation) wave of the inverter, the electromotive voltage due to the imbalance of the inductance can be detected as in the technique of Patent Document 2, and as a result, rotation occurs. Child position information can be obtained. Furthermore, with the technique of Patent Document 3, it is possible to make the drive waveform an ideal sine wave current.

In the technique described in Patent Document 4, as in the technique described in Patent Document 3, “neutral point potential” that is the potential of the connection point of the three-phase stator winding is detected to obtain position information. . When detecting the neutral point potential in synchronization with the PWM (pulse width modulation) wave of the inverter, constant reference potentials ((1/3) E, (2/3) E generated by dividing the power supply voltage E ) And the neutral point potential to obtain rotor position information. Also in the technique of Patent Document 4, it is possible to make the drive waveform an ideal sine wave current.

Among the techniques of Patent Documents 1 to 4, the techniques of Patent Document 3 and Patent Document 4 are useful as position detection means when the rotational speed of the motor, which is one of the problems of rotational position sensorless control, is low.

Furthermore, if the combination of the permanent magnet synchronous motor windings and inverters connected in a one-to-one manner is one system, and the number of systems consisting of combinations of windings and inverters is two or more for one permanent magnet synchronous motor. Even if one system fails, other systems can continue to operate. However, even in a multi-system permanent magnet synchronous motor drive system, it is necessary to obtain the position information of the rotor of the permanent magnet synchronous motor for each system.

JP 7-245981 A JP 2009-189176 A JP 2010-74889 A International Publication No. 2013/153656

In the technique of Patent Document 1, saliency is required for the rotor structure of the permanent magnet synchronous motor. If there is no saliency or less, the position detection sensitivity is lowered, and position estimation becomes difficult. Further, in order to detect with high sensitivity, it is necessary to increase the high frequency component to be injected or lower the frequency. As a result, rotation pulsation, vibration and noise increase, and harmonic loss of the permanent magnet synchronous motor increases.

In the technique of Patent Document 2, since an electromotive voltage generated in a non-energized phase of a three-phase winding is observed, the permanent magnet synchronous motor can be driven from a stopped state, but the drive current waveform is 120 degrees energized (rectangular wave) )become. Originally, it is more advantageous to drive a permanent magnet synchronous motor with a sinusoidal current in order to suppress rotation unevenness and harmonic loss, but sinusoidal driving is difficult in the technique of Patent Document 2.

In the techniques of Patent Document 3 and Patent Document 4, “neutral point potential” that is the potential of the connection point of the three-phase stator winding is detected to obtain position information. By detecting the neutral point potential in synchronization with the pulse voltage applied from the inverter to the motor, a potential change depending on the rotor position can be obtained. The position information can also be obtained by PWM (pulse width modulation) obtained by normal sine wave modulation as the voltage applied to the motor. Therefore, according to the techniques of Patent Document 3 and Patent Document 4, the problems in the techniques of Patent Document 1 and Patent Document 2 described above can be solved. However, the techniques of Patent Document 3 and Patent Document 4 have a problem that, when applied to a synchronous motor having a plurality of three-phase windings, a position estimation error occurs due to an offset included in a neutral point potential.

Therefore, the present invention provides a control device for a three-phase synchronous motor that can improve the detection accuracy of the rotor position even if an offset is included in the neutral point potential of the synchronous motor having a plurality of three-phase windings.

In order to solve the above problems, a control device for a three-phase synchronous motor according to the present invention includes a three-phase synchronous motor including a first three-phase winding and a second three-phase winding, and a first three-phase winding. The first inverter connected to the second inverter, the second inverter connected to the second three-phase winding, and the rotor position of the three-phase synchronous motor based on the neutral point potential of the first three-phase winding A first control unit that controls the first inverter based on the estimated rotor position, and a rotor position of the three-phase synchronous motor based on the neutral point potential of the second three-phase winding And a second control unit that controls the second inverter based on the estimated rotor position, wherein the first control unit is based on the DC voltage of the first inverter. Thus, the neutral point potential of the first three-phase winding is corrected.

According to the present invention, since the offset of the neutral point potential is corrected, the detection accuracy of the rotor position is improved.

Issues, configurations, and effects other than those described above will be clarified by the following description of embodiments.

An example of a PWM waveform and a neutral point potential waveform is shown. An example of a permanent magnet synchronous motor and the connection of a motor winding and an inverter are shown. The switching state of the inverter output voltage and the detected neutral point potential are shown. The result of the electromagnetic field analysis of the fluctuation of the neutral point potential of the system 1 when the motor is driven by the inverter of the system 1 is shown. It is a block diagram which shows the structure of the control apparatus of the three-phase synchronous motor which is Embodiment 1. FIG. The block diagram of the control part of the system | strain 1 is shown. It is a vector diagram showing a switching pattern of inverter output voltage, and a vector diagram showing a relationship between a rotor position and a voltage vector. The relationship between a permanent-magnet synchronous motor and a virtual neutral point circuit in the state where the voltage vector was applied is shown. 3 is a block diagram illustrating a configuration of a neutral point potential detection unit of system 1. FIG. The current / voltage in the winding of the system 1 when the voltage vector V (1, 0, 0) is output is shown. The neutral point potential with the offset corrected is shown. In the second embodiment, an example of a permanent magnet synchronous motor having a plurality of systems and the connection between the motor winding and the inverter are shown. It is a block diagram which shows the structure of the control apparatus of the three-phase synchronous motor which is Embodiment 2. FIG. The neutral point potential offset calculating means in the neutral point potential detection part with which the control part of the system | strain 1 in Embodiment 2 is provided is shown. It is a block diagram which shows the structure of the control apparatus of the three-phase synchronous motor which is Embodiment 3. FIG. The block diagram of the control part of the system | strain 1 is shown. It is a flowchart which shows the determination process which the detection position determination means in the system | strain 1 performs. FIG. 10 is a block diagram illustrating a configuration of a control unit of system 1 in a motor control device according to a fourth embodiment. It is a block diagram which shows the control apparatus structure of the three-phase synchronous motor which is Embodiment 5. FIG. The structure of the electric power steering apparatus which is Embodiment 6 is shown.

Hereinafter, embodiments of the present invention will be described. First, offsets generated in the neutral point potential will be described.

FIG. 1 shows an example of a PWM waveform and a neutral point potential waveform. The PWM pulse waveforms PVu, PVv, PVw are generated by comparing the three-phase voltage commands Vu *, Vv *, Vw * with the triangular wave carrier. The three-phase voltage commands Vu *, Vv *, and Vw * have a sinusoidal waveform, but can be regarded as a sufficiently lower frequency than the triangular wave carrier during low-speed driving. 1 can be regarded as a direct current.

The PWM pulse waves PVu, PVv, and PVw are repeatedly turned on and off at different timings. The voltage vectors in the figure have names such as V (0, 0, 1), but their subscripts (0, 0, 1) indicate the switch states of the U, V, and W phases, respectively. . That is, V (0, 0, 1) indicates PVu = 0 for the U phase, PVv = 0 for the V phase, and PVw = 1 for the W phase. Here, V (0, 0, 0) and V (1, 1, 1) are zero vectors in which the voltage applied to the motor is zero.

As shown in these waveforms, a normal PWM wave has two types of voltage vectors between the first zero vector V (0,0,0) and the second zero vector V (1,1,1). V (0,0,1) and V (1,0,1) are generated. That is, the voltage vector transition pattern “V (0,0,0) → V (0,0,1) → V (1,0,1) → V (1,1,1) → V (1,0, 1) → V (0,0,1) → V (0,0,0) ”as one cycle, this pattern is repeated. The voltage vectors used between the zero vectors are the same during the period in which the magnitude relationship of the three-phase voltage commands Vu *, Vv *, and Vw * does not change.

FIG. 3 shows the switching state of the inverter output voltage and the detected neutral point potential. The neutral point potential has eight voltage vectors: V (0,0,0), V (1,1,1), V (1,0,0), V (1,0,1), V (0 , 0, 1), V (0, 1, 1), V (0, 1, 0), V (1, 1, 0), V (0, 0, 0), V ( Detected for six voltage vectors other than 1,1,1). In the following, six voltage vectors V (1,0,0), V (1,0,1), V (0,0,1), V (0,1,1), V (0,1,1) The neutral point potentials detected for 0) and V (1, 1, 0) are VnA, VnB, VnC, VnD, VnE, and VnF, respectively, as shown in the figure.

The neutral point potential when a voltage vector other than these zero vectors is applied varies depending on the rotor position. Using this, the rotor position can be estimated. Here, when the change in the neutral point potential depending on the rotor position is small, the potential change is amplified using an amplification amplifier.

When an amplification amplifier is used, an offset error occurs due to variations in individual characteristics of the operational amplifier, resistor, and capacitor used in the amplification amplifier. On the other hand, in the technique of the above-described Patent Document 4, an offset error is accumulated in a memory in accordance with one electrical angle cycle, and the accumulated offset error is subtracted from the detected neutral point potential.

However, when the rotational position sensorless control using the neutral point potential at zero speed or extremely low speed is applied to a motor control device that drives one permanent magnet synchronous motor with two or more inverters, it is practical. There's a problem. As an example, a case where one permanent magnet synchronous motor is driven by two inverters will be described.

FIG. 2 shows an example of a permanent magnet synchronous motor having a plurality of systems, and the connection between the motor windings and the inverter. This motor is an 8-pole 12-slot motor having 8 poles and 12 slots. U-phase, V-phase, and W-phase windings are wound around the electromagnetic steel sheets in slots of the permanent magnet synchronous motor 4, which are grooves provided in a stator core made of laminated electromagnetic steel sheets. In system 1, inverter 1 and three-phase winding 41 (U1, U2, V1, V2, W1, W2) are connected, and in system 2, inverter 2 and three-phase winding 42 (U3, U4, V3, V4, W3, W4) are connected. Rotational position sensorless control is performed using neutral point potential Vn-m of system 1 and neutral point potential Vn-s of system 2.

At this time, since the three-phase winding 41 of the system 1 and the three-phase winding 42 of the system 2 are wound around the same electromagnetic steel plate, the phase of the system 1, the phase of the system 2, and the systems 1 and 2 Can be regarded as magnetically coupled. Therefore, mutual inductance components exist between the windings of each phase in the system 1, between the windings of each phase of the system 2, and between the windings of the system 1 and the windings of the system 2.

According to the study of the present inventor, such a mutual inductance causes an offset of the neutral point potential as will be described below.

FIG. 4 shows the result of electromagnetic field analysis of fluctuations in the neutral point potential Vn-m of the system 1 when the motor is driven by the inverter 31 of the system 1. The vertical axis represents the fluctuation of the neutral point potential by the ratio of the magnitude of the fluctuation of the neutral point potential to the magnitude of the power supply voltage (however, “%”).
The horizontal axis shows the rotor phase in terms of electrical angle.

As shown in FIG. 4, offsets are generated in neutral point potentials (FIG. 3), VnA, VnB, VnC, VnD, VnE, and VnF detected for six voltage vectors other than the zero vector. Depending on the voltage vector, the magnitude of the offset varies. For example, in VnA, a relatively large offset of 1.93% occurs with respect to the power supply voltage.

According to the inventor's study, the offset of the neutral point potential is caused by the difference in inductance between phases and the difference in mutual inductance between phases due to power supply voltage, magnetic asymmetry, and driving of multiple inverters. Fluctuates depending on. If the rotor position is estimated from the neutral point potential including such an offset error, a position estimation error occurs. For this reason, rotation irregularity and torque pulsation of the electric motor are generated, and vibration and noise are generated.

Each embodiment described below removes the influence of the offset of the neutral point potential as described above, and improves the estimation accuracy of the rotor position based on the neutral point potential.

Next, Embodiments 1 to 6 of the present invention will be described with reference to the drawings. In each figure, the same reference numerals indicate the same constituent elements or constituent elements having similar functions.
(Embodiment 1)
FIG. 5 is a block diagram showing a configuration of a three-phase synchronous motor control device (hereinafter referred to as “motor control device”) according to the first embodiment of the present invention.

The motor control device 3 drives and controls the permanent magnet synchronous motor 4 as a three-phase synchronous motor. The motor control device 3 includes a DC power source 5, an inverter 31 of the system 1 including the inverter main circuit 311 and the one-shunt current detector 312, an inverter 32 of the system 2 including the inverter main circuit 321 and the one-shunt current detector 322, and A permanent magnet synchronous motor 4 to be driven is provided.

The inverter 31 and the inverter 32 convert the DC power supplied from the DC power supply 5 into three-phase AC power and output it by controlling on / off of the semiconductor switching element. The permanent magnet synchronous motor 4 is driven by the three-phase AC power output from the inverter 31 and the inverter 32.

In the first embodiment, a MOSFET (Metal-Oxide-Semiconductor-Field-Effect-Transistor) is applied as a semiconductor switching element constituting the inverter main circuits 311 and 321. Moreover, the inverters 31 and 32 are voltage types, and generally a free-wheeling diode is connected to the semiconductor switching element in antiparallel. In the first embodiment, since the MOSFET built-in diode is used as the freewheeling diode, the freewheeling diode is not shown in FIG. In place of the MOSFET, an IGBT (Insulated Gate Bipolar Transistor) or the like may be applied. A freewheeling diode may be externally attached.

The permanent magnet synchronous motor 4 is a multi-winding motor (two windings in the first embodiment), and includes a three-phase winding 41 and a three-phase winding 42 provided on the same stator. In the first embodiment, as shown in FIG. 2 described above, in one stator, a region where the three-phase windings 41 (U1, U2, V1, V2, W1, W2) of the system 1 are provided, The region where the three-phase three-phase winding 42 (U3, U4, V3, V4, W3, W4)) is provided is divided. That is, in a circular cross section perpendicular to the rotation axis, the three-phase winding 41 of the system 1 is provided in one semicircular part with a specific diameter as a boundary, and the three-phase winding 42 of the system 2 is provided in the other semicircular part. Is provided. If this specific diameter is viewed in the horizontal direction (or vertical direction) in the cross section of FIG. 2, the region where the three-phase winding 41 is provided and the region where the three-phase winding 42 is provided are vertically (or left and right). ). Therefore, such a structure is hereinafter referred to as “upper and lower (or left and right) separation structure”.

The combination of the number of poles and the number of slots is 8 poles and 12 slots as shown in FIG. If the same stator is provided with multiple systems of three-phase windings and an inverter is connected to each of the three-phase windings, the combination of the number of poles and the number of slots is appropriately set according to the desired motor performance. Good.

The inverter 31 of the system 1 includes an output pre-driver 313 in addition to the inverter main circuit 311 and the one-shunt current detector 312.

The inverter main circuit 311 is a three-phase full bridge circuit composed of six semiconductor switching elements Sup1 to Swn1.

The one shunt current detector 312 detects the supply current I0-m (DC bus current) to the inverter main circuit 311 of the system 1.

The output pre-driver 313 is a driver circuit that directly drives the semiconductor switching elements Sup1 to Swn1 of the inverter main circuit 311.

The inverter 32 of the system 2 includes an output pre-driver 323 in addition to the inverter main circuit 321 and the one-shunt current detector 322.

The inverter main circuit 321 is a three-phase full bridge circuit composed of six switching elements Sup2 to Swn2.

The one shunt current detector 322 detects the supply current I0-s (DC bus current) to the inverter main circuit 321 of the system 2.

The output pre-driver 323 is a driver that directly drives the semiconductor switching elements Sup2 to Swn2 of the inverter main circuit 321.

The three-phase current flowing through the three-phase winding 41 is measured by the so-called one-shunt method based on the DC bus current I0-m detected by the one-shunt current detector 312. Similarly, based on the DC bus current I0-s detected by the one-shunt current detector 322, the three-phase current flowing through the three-phase winding 42 is measured. Since the one-shunt method is a known technique, a detailed description is omitted.

DC power supply 5 supplies DC power to inverter 31 of system 1 and inverter 32 of system 2. Note that DC power may be supplied to the inverter 31 and the inverter 32 by separate DC power supplies.

The control unit 61 of the system 1 estimates and calculates the rotor position (θd−m) based on the neutral point potential Vn−m of the three-phase winding 41, and outputs the pre-driver based on the estimated rotor position. A gate command signal to be given to 313 is created.

The control unit 62 of the system 2 estimates and calculates the rotor position (θd−s) based on the neutral point potential Vn−s of the three-phase winding 42, and outputs the predriver based on the estimated rotor position. A gate command signal to be given to H.323 is created.

FIG. 6 shows a block diagram of the control unit 61 of the system 1. In the control unit 61, so-called vector control is applied. Note that the configuration of the control unit 62 of the system 2 is the same as that of the control unit 61, and thus the description thereof is omitted.

As shown in FIG. 6, the control unit 61 of the system 1 includes a q-axis current command generation unit (Iq * generation unit) 611, a d-axis current command generation unit (Id * generation unit) 612, a subtraction unit 613a, and a subtraction unit 613b. , D-axis current control means (IdACR) 614a, q-axis current control means (IqACR) 614b, dq reverse conversion means 615, PWM generation means 616, current reproduction means 617, dq conversion means 618, sample / hold means (S / H) Circuit) 619, speed calculation means 620, pulse shift means 621, neutral point potential detection unit 622, and rotational position estimation unit 623. With this configuration, the control unit 61 operates so that the permanent magnet synchronous motor 4 generates torque according to the q-axis current command Iq * and the d-axis current command Iq *.

The Iq * generating means 611 generates a q-axis current command Iq * corresponding to the motor torque. The Iq * generating means 611 normally generates a q-axis current command Iq * so that the rotational speed of the permanent magnet synchronous motor 4 becomes a predetermined value while observing the actual speed ω1. The q-axis current command Iq *, which is the output of the Iq * generating means 611, is output to the subtracting means 613b.

The Id * generating means 612 generates a d-axis current command Id * corresponding to the exciting current of the permanent magnet synchronous motor 4. The d-axis current command Id * that is the output of the Id * generating means 612 is output to the subtracting means 613a.

The subtracting unit 613a includes a d-axis current command Id * output from the Id * generating unit 612 and a d-axis current Id output from the dq converting unit 618, that is, a three-phase current (Iuc, Ivc, The deviation from the d-axis current Id obtained by dq conversion of Iwc) is obtained.

The subtracting unit 613b includes a q-axis current command Iq * output from the Iq * generating unit 611 and a q-axis current Iq output from the dq converting unit 618, that is, a three-phase current (Iuc, Ivc, A deviation from the q-axis current Iq obtained by dq conversion of Iwc) is obtained.

The IdACR 614a calculates the d-axis voltage command Vd * on the dq coordinate axis so that the d-axis current deviation calculated by the subtracting unit 613a becomes zero. Further, the IqACR 614b calculates the q-axis voltage command Vq * on the dq coordinate axis so that the q-axis current deviation calculated by the subtracting unit 613b becomes zero. The d-axis voltage command Vd * that is the output of IdACR 614a and the q-axis voltage command Vq * that is the output of IqACR 614b are output to dq inverse conversion means 615.

The dq reverse conversion means 615 converts the voltage commands Vd *, Vq * of the dq coordinate (magnetic flux axis-magnetic flux axis orthogonal axis) system into voltage commands Vu *, Vv *, Vw * on the three-phase AC coordinates. The dq inverse conversion means 615 is configured to output the voltage commands Vu *, Vq * of the three-phase AC coordinate system based on the voltage commands Vd *, Vq * and the rotor position θd−m output from the rotational position estimating unit 623 (FIG. 6) of the system 1. Vv * and Vw * are calculated. The dq inverse conversion unit 615 outputs the calculated Vu *, Vv *, Vw * to the PWM generation unit 616.

The PWM generation means 616 outputs a PWM (Pulse Width Modulation) signal for controlling the power conversion operation of the inverter main circuit 311 of the system 1. The PWM generation unit 616 compares the three-phase AC voltage command with a carrier signal (for example, a triangular wave) based on the three-phase AC voltage commands Vu *, Vv *, and Vw *, thereby generating a PWM signal (FIG. 9, which will be described later). PVu, PVv, and PVw at 10 and 14 are generated. The PWM signal output from the PWM generation means 616 is input to the output pre-driver 313 (FIG. 4) as a gate command signal via the pulse shift means 621 described later and also input to the sample / hold means 619.

The current reproduction means 617 reproduces the three-phase current (Iuc, Ivc, Iwc) flowing through the three-phase winding 41 from the DC bus current I0-m output from the inverter main circuit 311 to the one-shunt current detector 312. The reproduced three-phase current (Iuc, Ivc, Iwc) is output from the current reproduction means 617 to the dq conversion means 618.

The dq conversion means 618 converts the three-phase current (Iuc, Ivc, Iwc) into Id, Iq on the dq coordinate that is the rotation coordinate axis. The converted Id and Iq are used for calculating the deviation from the current command in the subtracting means 613a and 613b, respectively.

The sample / hold means 619 samples the value of the DC bus current I0-m at the timing of acquiring the phase current information, that is, the timing at which the PWM pulse is switched based on the PWM signal input via the pulse shift means 621, The sampled value is output to the current reproduction means 617 while being held. Since specific sampling timing is a known technique, a detailed description thereof will be omitted.

The speed calculation means 620 calculates the rotational speed ω1 of the permanent magnet synchronous motor from the rotor position θd−m that is the estimated value of the rotational position estimation unit 623. The calculated rotational speed ω1 is output to the Iq * generating means 611 and used for current control on the axis (q axis) orthogonal to the magnetic flux axis (d axis).

The pulse shift means 621 includes a PWM generator 616 for increasing the current passing time of the phase current flowing in the DC bus current at the sampling timing so that accurate phase current information can be obtained by the sample / hold means 619. Are shifted to the sample / hold means 619 and the neutral point potential detector 622. Since pulse shift is a known technique, detailed description thereof is omitted.

The neutral point potential detection unit 622 detects the neutral point potential Vn-m of the three-phase winding that is star-connected, and removes the offset from the detected value of Vn-m, as will be described later. The neutral point potential value Vn−m ′ is output.

The rotational position estimation unit 623 estimates and calculates the rotor position θd−m for the three-phase winding 41 based on the neutral point potential value Vn−m ′ from which the offset input from the neutral point potential detection unit 622 is removed. . As described above, the neutral point potential when a voltage vector other than the zero vector is applied varies depending on the rotor position. Using this, the rotor position is estimated. Since specific means for estimating the rotor position from the neutral point potential is a known technique (see, for example, Patent Document 3 and Patent Document 4 described above), detailed description thereof is omitted.

In the first embodiment, the control unit 61 of the system 1 is configured by a single microcomputer. Moreover, the control part 62 of the system | strain 2 is comprised by another one microcomputer. The neutral point of the three-phase winding 41 and the neutral point of the three-phase winding 42 are respectively electrically connected to the control microcomputer in the system 1 and the control microcomputer in the system 2 by wiring or the like. The Thus, the neutral point potential detection unit (622 in FIG. 6) detects the neutral point potential of the three-phase winding. Alternatively, as described later (FIG. 8), the neutral point potential may be detected using a virtual neutral point circuit.

Furthermore, each of the inverter main circuit 311, the output predriver 313, the inverter main circuit 321, and the output predriver 323 may be configured by an integrated circuit device. Further, each of the inverter 31 and the inverter 32 may be constituted by an integrated circuit device. As a result, the motor control device can be greatly reduced in size. In addition, the motor control device can be easily mounted on various electric devices, and the various electric devices can be downsized.

Next, the basic operation of this motor drive system will be described.

In the first embodiment, vector control generally known as control means for linearizing the torque of the synchronous motor is applied.

The principle of the vector control technique is a method of independently controlling the current Iq contributing to the torque and the current Id contributing to the magnetic flux on the rotation coordinate axis (dq coordinate axis) based on the rotor position of the motor. The d-axis current control unit 614a, the q-axis current control unit 614b, the dq reverse conversion unit 615, the dq conversion unit 618, and the like in FIG. 6 are main parts for realizing this vector control technique.

In the control unit 61 of the system 1 in FIG. 6, the current command Iq * corresponding to the torque current is calculated by the Iq * generation means 611, and the current command Iq * and the actual torque current Iq of the permanent magnet synchronous motor 4 are calculated. Current control is performed so as to match.

The current command Id * is normally given as “zero” if it is a non-salient permanent magnet synchronous motor. On the other hand, in a salient pole structure permanent magnet synchronous motor or field weakening control, a negative command may be given as the current command Id *.

The three-phase current of the permanent magnet synchronous motor is directly detected by a current sensor such as a CT (Current Transformer), or is reproduced and calculated in the controller based on the DC bus current as in the first embodiment. To do. In the first embodiment, a three-phase current is reproduced and calculated from the DC bus current I0-m of the system 1 and the DC bus current I0-s of the system 2. For example, the control unit 61 shown in FIG. 6 samples the current value of the DC bus current I0-m by operating the S / H unit 619 at a timing corresponding to the PWM signal phase-shifted by the pulse shift unit 621. By holding, the current value of the DC bus current I0-m including information on the three-phase current is acquired. Then, the three-phase current (Iuc, Ivc, Iwc) is reproduced and calculated by the current reproducing means 617 from the acquired current value. In addition, since the concrete means of reproduction calculation is a well-known technique, detailed description is abbreviate | omitted.

In the first embodiment, the reference rotor position in the rotating coordinate system is estimated by the rotating position estimation unit based on the neutral point potential of the three-phase winding. For example, in the control unit 61 of the system 1, the rotational position estimation unit 623 is configured to detect the three-phase winding 41 based on the neutral point potential Vn−m ′ from which the offset has been removed. The rotor position θd−m is estimated for. In the system 2 as well, the rotor position is similarly estimated for the three-phase winding 42.

Hereinafter, the means for estimating the rotor position from the neutral point potential in the first embodiment will be described with the system 1 as a representative.

First, the neutral point potential fluctuation will be described.

The output potential of each phase of the inverter 31 is set by the on / off state of the upper semiconductor switching elements (Sup1, Svp1, Swp1) or the lower semiconductor switching elements (Sun1, Svn1, Swn1) of the inverter main circuit 311. In each of these semiconductor switching elements, if one of the upper side and the lower side is in an on state, the other is in an off state. That is, in each phase, the upper and lower semiconductor switching elements are complementarily turned on / off. Therefore, the output voltage of the inverter 31 has eight switching patterns in total.

FIG. 7 is a vector diagram (left diagram) showing the switching pattern of the inverter output voltage and a vector diagram (right diagram) showing the relationship between the rotor position (phase) θd and the voltage vector.

Each vector has a name such as V (1,0,0). In this vector notation, the state in which the upper semiconductor switching element is on is represented by “1”, and the state in which the lower semiconductor switching element is on is represented by “0”. The numbers in parentheses indicate the switching states in the order of “U phase, V phase, W phase”. The inverter output voltage can be expressed using eight voltage vectors including two zero vectors (V (0,0,0), V (1,1,1)). By combining these eight voltage vectors, a sinusoidal current is supplied to the permanent magnet synchronous motor 4.

As shown in FIG. 7 (right diagram), the rotor position (phase) θd is defined with the reference of the rotor position of the permanent magnet synchronous motor 4 as the U-phase direction. The dq coordinate axis in the rotation coordinate is the counter-clockwise rotation with the direction of the magnet magnetic flux Φm as the d-axis direction. Note that the q-axis direction is a direction orthogonal to the d-axis direction.

Here, when θd = 0 °, the induced voltage vector Em is in the q-axis direction, so that it is close to the voltage vectors V (1, 0, 1) and V (0, 0, 1). positioned. In this case, the permanent magnet synchronous motor 4 is driven mainly using the voltage vectors V (1, 0, 1) and V (0, 0, 1). Note that voltage vectors V (0,0,0) and V (1,1,1) are also used, but these are zero vectors.

FIG. 8 shows the relationship between the permanent magnet synchronous motor 4 and the virtual neutral point circuit 34 when the voltage vector is applied. Here, Lu, Lv, and Lw are the inductance of the U-phase winding, the inductance of the V-phase winding, and the inductance of the W-phase winding, respectively. The applied voltage vectors are the above-described voltage vectors V (1, 0, 1) (left figure) and V (0, 0, 1) (right figure).

The neutral point potential Vn0 shown in FIG. 8 can be calculated as follows. The neutral point potential of the virtual neutral point circuit is used as the reference potential.

When the voltage vector V (1, 0, 1) is applied, it is calculated by the equation (1).

When the voltage vector V (0, 0, 1) is applied, it is calculated by the equation (2).

Here, the notation “//” is the total inductance value of the parallel circuit of two inductances, and for example, “Lu // Lw” is expressed by Equation (3).

If the three-phase winding inductances Lu, Lv, and Lw are all equal, the neutral point potential Vn0 is zero according to the equations (1) and (2). However, in practice, there is a considerable difference in inductance due to the influence of the permanent magnet magnetic flux distribution of the rotor. That is, the magnitudes of the inductances Lu, Lv, and Lw vary depending on the position of the rotor, and there are differences in the magnitudes of Lu, Lv, and Lw. For this reason, the magnitude of the neutral point potential Vn0 changes according to the rotor position.

FIG. 1 shows the state of pulse width modulation using a triangular wave carrier, the voltage vector at that time, and the change of the neutral point potential. Here, the triangular wave carrier is a signal serving as a reference for converting the magnitude of the three-phase voltage commands Vu *, Vv *, and Vw * into a pulse width, and the triangular wave carrier and the three-phase voltage commands Vu *, Vv. A PWM pulse is created by comparing the magnitude relationship between * and Vw *. As shown in FIG. 1, the rise / fall of the PWM pulse changes at the time when the magnitude relationship between each voltage command Vu *, Vv *, Vw * and the triangular wave carrier changes. At the same time, a non-zero neutral point potential Vn0 is detected.

As shown in FIG. 1, the neutral point potential Vn0 hardly fluctuates except at the rising / falling time of the PWM pulse. This indicates that the difference in the sizes of the three-phase winding inductances Lu, Lv, and Lw generated according to the rotor position is small. In contrast, when the PWM pulse rises / falls, that is, when a voltage vector other than the zero vector (V (1, 0, 1) and V (0, 0, 1) in FIG. 1) is applied. Since the change rate of the motor current is increased, a relatively large change in the neutral point potential Vn0 is detected even if the difference in the magnitude of the inductance is small. Therefore, if the neutral point potential is observed in synchronization with the PWM pulse signals PVu, PVv, and PWw, the fluctuation of the neutral point potential can be detected with high sensitivity.

Next, means for estimating the rotor position from the detected neutral point potential will be described.

Since the neutral point potential Vn0 changes periodically according to the rotor position (see, for example, Patent Document 3 and Patent Document 4 described above), the relationship between the rotor position and the neutral point potential Vn0 is measured or measured in advance. Simulation is performed to obtain map data, table data, or a function indicating the relationship between the rotor position and the neutral point potential Vn0. The rotor position is estimated from the detected neutral point potential using such map data, table data, or function.

Further, the neutral point potential detected for two types of voltage vectors (in FIG. 1, V (1, 0, 1) and V (0, 0, 1)) is regarded as a three-phase alternating current amount (for two phases). Then, the phase amount is calculated using coordinate transformation (three-phase two-phase transformation), and this phase amount is used as the estimated value of the rotor position. In addition, since this means is based on a well-known technique (for example, refer the above-mentioned patent document 4), detailed description is abbreviate | omitted.

The rotational position estimation unit 623 (FIG. 6) of the system 1 uses the estimation means as described above, based on the neutral point potential Vn−m ′ output from the neutral point potential detection unit 622 (FIG. 6). The position θd−m is estimated. These estimation means are appropriately selected according to the desired position detection accuracy and the performance of the control microcomputer. The same applies to the system 2.

Hereinafter, the neutral point potential detection unit 622 (FIG. 6) according to the first embodiment will be described.

The neutral point potential detector 622 of the system 1 detects the neutral point potential Vn−m of the three-phase winding 41, removes the offset from the detected value of Vn−m, and removes the neutral point potential from which the offset is removed. The value Vn−m ′ is output. The system 2 is also provided with a neutral point potential detection unit, but since it is the same as the system 1, description thereof is omitted.

The neutral point potential detector 622 of the system 1 will be described.

FIG. 9 is a block diagram illustrating a configuration of the neutral point potential detection unit 622 of the system 1.

As shown in FIG. 9, the neutral point potential detection unit 622 includes a neutral point potential offset calculation unit 624 and a potential detection unit 625.

The neutral point potential offset calculating means 624 calculates the neutral point potential offset from the gate command signal that is the output of the pulse shift means 621 (FIG. 6) and the power supply voltage E that is the output of the DC power supply 5 (FIG. 5).

Next, the function of the neutral point potential offset calculation means 624 will be specifically described.

2 has the structure in which the three-phase windings of the stator are vertically separated (or left and right) separated. In the three-phase winding 41, the windings are intensively wound around the respective teeth of the stator in the order of W1 → U1 → V1 → W2 → U2 → V2 counterclockwise in FIG. Therefore, in the three-phase winding 41, a difference occurs in the mutual inductance Muv1 between the UV phases, the mutual inductance Mvw1 between the VW phases, and the mutual inductance Mwu1 between the WU phases. For this reason, an offset of the neutral point potential occurs. Further, the offset of the neutral point potential varies depending on the switching pattern of the inverter 31 and the power supply voltage E.

As shown in FIG. 4 described above, among the neutral point potentials, for example, VnA detected when V (1, 0, 0) is output has a voltage offset of 1.93% in the positive direction. VnD detected when V (0, 1, 1) is output has a voltage offset of 1.93% in the negative direction. As shown in FIG. 4, the voltage offset differs in direction and magnitude for each voltage vector and presents a complicated pattern. Therefore, in the first embodiment, such an offset of the neutral point potential is obtained by calculation as follows.

In order to calculate the offset, a three-phase voltage equation of the motor as shown in Equation (4) is used.

Here, Vun1, Vvn1, and Vwn1 are the phase voltage between UN, the phase voltage between VN, and the phase voltage between WN, respectively. Lu, Lv, Lw, Muv, Mvw, and Mwu are U phase self-inductance, V phase self inductance, W phase self inductance, UV phase mutual inductance, VW phase mutual inductance, and WU phase mutual inductance, respectively. It is. Iu1, Iv1, and Iw1 are a U-phase current, a V-phase current, and a W-phase current, respectively. eu, ev, and ew are a U-phase electromotive voltage, a V-phase electromotive voltage, and a W-phase electromotive voltage, respectively. R is a winding resistance. p is a differential operator (d / dt).

FIG. 10 shows the current and voltage in the three-phase winding 41 of the system 1 when the voltage vector V (1, 0, 0) is output. Note that “E” in the figure indicates the power supply voltage from the DC power supply 5 applied to the three-phase winding 41 via the inverter main circuit 311 (FIG. 5).

In the case of FIG. 10, the U-phase upper arm of the inverter main circuit 311 is on, the V-phase and W-phase lower arms are on, and the detected neutral point potential is VnA. Note that the neutral point potential in FIG. 10 represents the reference potential as the ground potential (the low potential side of the DC power supply 5 (see FIG. 5)) using a virtual neutral point circuit (see FIG. 8). From FIG. 10, Equation (5) showing the relationship between the phase voltages Vun1, Vvn1, Vwn1 and the neutral point potential VnA is obtained.

The neutral point potential VnA when the voltage vector V (1, 0, 0) is output is calculated from the expressions (1) and (5) and the known relational expression “Iu1 + Iv1 + Iw1 = 0” regarding the three-phase equilibrium current. it can. However, assuming that the current flowing through the three-phase winding 41 is small and the motor is stopped or rotating at an extremely low speed, the first and third terms on the right side of the equation (4) are ignored.

For the neutral point potentials VnB, VnC, VnD, VnE, and VnF (see FIG. 3) when other voltage vectors are output, the same relational expression as the expression (5) related to VnA is obtained. From the equation (1), each neutral point potential can be calculated.

The neutral point potentials VnA, VnB, VnC, VnD, VnE, and VnF calculated as described above are expressed by Expression (6). However, it is assumed that the self-inductances (Lu, Lv, Lw) of the respective phases are equal and L.

In equation (6), when the mutual inductance is zero, the neutral point potential calculated is zero as described above with reference to FIG. Further, as described above, the rise / fall time of the PWM pulse, that is, a voltage vector other than the zero vector (V (1, 0, 1) and V (0, 0, 1) in FIG. 1) is applied. Since the rate of change of the motor current increases, the neutral point potential is detected. Therefore, in this case, the neutral point potential is detected without an offset.

On the other hand, when there is mutual inductance, a neutral point potential represented by the equation (6) is calculated. When the rate of change of the motor current increases at the rising / falling time of the PWM pulse, the neutral point potential varies from the neutral point potential represented by the equation (6). That is, the neutral point potential represented by the equation (6) corresponds to a voltage offset included in the detected neutral point potential.

In the configuration of the three-phase winding 41 of this embodiment (see FIG. 2), since the magnitudes of Mvw and Mwu are different, there is a non-zero offset for each of the six voltage vectors other than the zero vector. Calculated (see FIG. 4).

Thus, equation (6) represents the neutral point potential offset when each voltage vector is output. As indicated by equation (6), the neutral point potential offset depends on the power supply voltage, the self-inductance of the winding, and the mutual inductance.

The neutral point potential offset calculating means 624 shown in FIG. 9 calculates the offset according to the voltage vector, that is, the switching pattern, based on the equation (6). Based on the gate command signal, the neutral point potential offset calculating means 624 determines which voltage vector corresponds to the switching pattern, that is, which of VnA to VnF in the equation (6) is to be calculated. .

The power supply voltage E used in Equation (6) is acquired by taking in information from the DC power supply 5 side or detecting an input voltage on the DC side of the inverter main circuit 311. The power supply voltage E may be taken out by direct AD conversion, or may be taken out after being divided and AD converted. Also, digital or analog filters or average value processing may be used.

The values of mutual inductance and self-inductance in Equation (6) are obtained by actual measurement or electromagnetic field analysis. Therefore, in the equation (6) set in the neutral point potential detection unit 622, the mutual inductance and the self-inductance are given constant values in advance, and the power supply voltage E, that is, the DC voltage of the inverter becomes a variable. That is, the neutral point potential detection unit 622 corrects the neutral point potential Vn−m based on the direct current voltage using Equation (6).

As shown in FIG. 9, in the neutral point potential detecting unit 622, the offset calculated by the neutral point potential offset calculating means 624 is subtracted from the neutral point potential Vn−m detected in the three-phase winding 41. The Then, the potential detection unit 625 outputs the voltage value from which the offset is removed from Vn−m as described above as the neutral point voltage detection value Vn−m ′.

FIG. 11 shows a neutral point potential in which the offset is corrected by the above-described means with respect to the neutral point potential shown in FIG. As shown in FIG. 11, the offset (maximum 1.93%) in FIG. 4 is removed in any of the neutral point potentials VnA to VnA.

As described above, the neutral point potential detection unit 622 is configured to detect the neutral point potential (Vn) detected in the three-phase winding based on the mutual inductance and self-inductance in the three-phase winding, the power supply voltage, and the gate command signal. -M) is corrected so that the offset is removed. By estimating the position of the rotor based on the neutral point potential value from which the offset has been removed in this way, the estimation accuracy of the rotor position is improved.

In the first embodiment, the neutral point potential detection unit 622 always corrects the neutral point potential offset according to each switching pattern by using the neutral point potential offset calculation means 624. Thereby, since the influence of the offset in the estimation of the rotor position based on the neutral point potential is always compensated, it is possible to estimate the rotor position with high accuracy.

In the first embodiment, the mutual inductance between phases in the three-phase winding 41 is used for offset correction. However, the mutual inductance between the three-phase winding 41 and the three-phase winding 42 may be used. . Thereby, in the case where the magnetic interference between the three-phase winding 41 and the three-phase winding 42 increases according to the configuration of the three-phase winding in the stator, the accuracy of estimating the rotor position is improved. In this case, a three-phase voltage equation that takes into account the mutual inductance between the three-phase winding 41 and the three-phase winding 42 is used.

In the above formula (6), the current dependency of the inductance is not particularly considered. On the other hand, the self-inductance and mutual inductance may be calculated based on the dq-axis current (Id, Iq) output from the dq conversion means 618 (FIG. 6) in consideration of the current dependency of the inductance.

As described above, according to the first embodiment, when the rotor position is estimated based on the neutral point potential of the three-phase winding, in the three-phase winding based on the power supply voltage, that is, the DC voltage of the inverter. Since the detected neutral point potential is corrected so that the offset depending on the mutual inductance in the three-phase winding is removed, the estimation accuracy of the rotor position is improved even if the neutral point potential has an offset can do. Since the offset is removed by the calculation function of the control unit, the estimation accuracy of the rotor position can be improved without increasing the number of parts of the motor control device.

Further, according to the first embodiment, position sensorless driving at a very low speed can be realized in a motor driving system using a three-phase synchronous motor in which the mutual inductance of each phase is different and the neutral point potential offset is large. .

Further, according to the first embodiment, when the three-phase synchronous motor is controlled based on the neutral point potential, the structure having few transitions between the windings as shown in FIG. 2 as the three-phase synchronous motor, that is, relatively large A permanent magnet synchronous motor that can generate an offset can be employed. Such a motor can shorten the connection length of the neutral point potential and is easy to assemble, so that the motor cost can be reduced.

The correction of the neutral point potential offset as described above is not limited to the case where the two three-phase windings provided in one stator are driven by two inverters, but one three-position provided in one stator. It can also be applied when the phase winding is driven by one inverter, or when three or more three-phase windings provided in one stator are driven by the same number of inverters, and the same effect can be obtained. .
(Embodiment 2)
Next, Embodiment 2 of the present invention will be described with reference to FIGS. Note that differences from the first embodiment will be mainly described.

FIG. 12 shows an example of a permanent magnet synchronous motor having a plurality of systems and the connection between the windings of the motor and the inverter in the second embodiment.

As shown in FIG. 12, the permanent magnet synchronous motor 4 in the second embodiment is a multi-winding (two-winding) motor as in the case of the first embodiment (FIG. 2). Three-phase windings (three-phase windings 41a and 42a) are provided. The three-phase winding 41a of the system 1 includes U-phase windings U1 and U2, V-phase windings V2 and V3, and W-phase windings W1 and W4. The three-phase winding 42a of the system 2 includes U-phase windings U3 and U4, V-phase windings V1 and V4, and W-phase windings W2 and W3. The three-phase winding 41a and the three-phase winding 42a are a pair of two phase windings of each system, and are alternately provided in the circumferential direction of a circular cross section perpendicular to the rotation axis direction of the stator. That is, in the permanent magnet synchronous motor 4 according to the second embodiment, unlike the first embodiment, the region where the three-phase winding 41a of the system 1 is provided and the region where the three-phase winding 42a of the system 2 is provided are complicated. It is out. Therefore, such a structure is hereinafter referred to as a “star structure”.

In the case of a three-phase winding having a star structure as shown in FIG. 12, due to the symmetry of each phase, the mutual inductance between the phases is almost equal in the same system. Therefore, as can be seen from Mvw = Mwu in the above equation (6), if only the mutual inductance in one system is considered, it is considered that the offset of the neutral point potential does not occur.

However, in the second embodiment, the magnetic coupling between the three-phase winding 41a of the system 1 and the three-phase winding 42a of the system 2 is strengthened due to the star structure of the three-phase winding. The mutual inductance between the three-phase winding 41a and the three-phase winding 42a of the system 2 increases. For this reason, an offset occurs in the neutral point potential. That is, an offset occurs in the neutral point potential of the three-phase winding 42a according to the switching pattern output from the inverter connected to the three-phase winding 41a, and the switching output from the inverter connected to the three-phase winding 42a. Depending on the pattern, an offset occurs in the neutral point potential of the three-phase winding of the three-phase winding 41a.

FIG. 13 is a block diagram showing a configuration of a three-phase synchronous motor control device (hereinafter referred to as “motor control device”), which is Embodiment 2 of the present invention.

As shown in FIG. 13, the motor control device of the second embodiment includes a control unit communication unit 63 that controls communication between the control unit 61a of the system 1 and the control unit 62a of the system 2. The control unit 61a and the control unit 62a have functions similar to those of the control unit (61 and 62 in FIG. 5) in the first embodiment, and further communicate with each other via the control unit communication unit 63 so that information in the own system is obtained. While being able to send to other systems, the information regarding other systems can be acquired. Thereby, the control unit 61a acquires the gate signal output from the control unit 62a to the output pre-driver 323, that is, information indicating the switching pattern output from the inverter 32 of the system 2. Similarly, the control unit 62a also acquires information indicating a switching pattern output from the inverter 31 of the system 1.

FIG. 14 shows the neutral point potential offset calculating means 624a in the neutral point potential detecting unit provided in the control unit 61a of the system 1 in the second embodiment (neutral point potential offset calculating means in the first embodiment (FIG. 9)). 624). In addition, since the structure of the control part 62a of the system | strain 2 is the same as that of the control part 61a, description is abbreviate | omitted about the control part 62a.

As shown in FIG. 14, the neutral point potential offset calculation means 624 a includes a gate command signal for the system 1, a gate command signal for the system 2 acquired from the control unit 62 a by the control unit communication unit 63, a power supply voltage E, Based on the above, an offset generated at the neutral point potential of the three-phase winding 41a of the system 1 is calculated according to the switching pattern of the inverter 31 of the system 1 and the switching pattern of the inverter 32 of the system 2. The other configuration of the neutral point potential detection unit of the second embodiment (corresponding to the neutral point potential detection unit 622 of the first embodiment (FIG. 9)) is the same as that of the first embodiment (FIG. 9).

In the second embodiment, as in the first embodiment, the offset is the three-phase voltage equation of the motor (corresponding to the above equation (3)) and the relationship between the phase voltage and the offset in each switching pattern (the above equation ( 4)) and the relational expression of the three-phase balanced current. However, in the three-phase voltage equation of the own system, in addition to the self-inductance and the mutual inductance of the own system, the mutual inductance between the winding of the own system and the winding of the other system is considered. Therefore, the three-phase voltage equation includes currents of other systems in addition to the phase current of the own system. Further, the offset is obtained from the three-phase voltage equation regarding the own system and the other system, the relationship between the phase voltage and the offset, and the relational expression of the three-phase balanced current according to the switching pattern of the own system and the other system. Note that the obtained offset includes the power supply voltage E as a parameter, as in the above-described equation (6).

The neutral point potential offset calculating means 624a includes a combination of the switching pattern of the system 1 and the switching pattern of the system 2, and the offset of the neutral point potential of the three-phase winding 41a of the system 1 obtained as described above. Expressions or data indicating the correspondence between these are preset. The neutral point potential offset calculation means 624a uses such a formula or data to combine the combination of the switching patterns of the system 1 and the system 2 indicated by the gate command signals of the system 1 and the system 2 to be input and the power supply voltage E to be input. Based on the above, the neutral point potential offset of the three-phase winding 41a of the system 1 is calculated.

As described above, according to the second embodiment, a permanent magnet synchronous motor having two three-phase windings is driven by two inverters, and the rotor position is determined based on the neutral point potential of the three-phase winding. The neutral point potential detected in the three-phase winding is calculated based on the self-inductance and mutual inductance of the three-phase winding in the own system, the mutual inductance between the own system and the other system, and the power supply voltage. Since the correction is made so that the offset is removed, the estimation accuracy of the rotor position can be improved even if the neutral point potential has the offset. Since the offset is removed by the calculation function of the control unit, the estimation accuracy of the rotor position can be improved without increasing the number of parts of the motor control device.

Further, according to the second embodiment, position sensorless driving at an extremely low speed can be realized in a motor driving system that drives a permanent magnet synchronous motor having two systems of three-phase windings by two systems of inverters.

In addition, if the control part 61a and the control part 62a are comprised with the same microcomputer, even if the control part communication part 63 is abbreviate | omitted, an offset can be corrected similarly.

Further, the offset correction means in the second embodiment is not limited to the permanent magnet synchronous motor having the same mutual inductance in one system as shown in FIG. 12, but there is a mutual inductance difference in one system, and between the systems. The present invention can also be applied to a permanent magnet synchronous motor in which mutual inductance exists.
(Embodiment 3)
In the third embodiment, the rotor position estimation based on the neutral point potential as described above and the rotational position detection using a rotational position detector (for example, Hall IC, resolver, encoder, GMR sensor) are used in combination. Normally, motor control is executed based on the rotor position detected by the rotational position detector. Further, an abnormality of the rotational position detector is determined based on the estimated rotational position based on the neutral point potential. When it is determined that the rotational position detector is abnormal, motor control is executed based on the estimated rotational position based on the neutral point potential. Accordingly, even if a malfunction such as a failure or a signal abnormality occurs in the rotational position detector, the motor control can be continued by the estimated rotor position, so that the reliability of the motor control device is improved.

Hereinafter, Embodiment 3 of the present invention will be described with reference to FIGS. 15 to 17. Note that differences from the first embodiment will be mainly described.

FIG. 15 is a block diagram showing a configuration of a control device for a three-phase synchronous motor (hereinafter referred to as “motor control device”), which is Embodiment 3 of the present invention.

15, in addition to the configuration of the first embodiment (FIG. 5), the rotational position detectors 411 and 412 are provided in the system 1, and the rotational position detectors 421 and 422 are provided in the system 2. In Embodiment 4, the reliability of rotational position detection by the rotational position detector is improved by redundantly providing a plurality of rotational position detectors in each system. As will be described later, highly accurate motor control can be maintained by determining whether or not the rotational position detectors 421 and 422 are abnormal and using the rotor position output by the normal rotational position detector.

Further, in the system 1, the rotor positions θd-11 and θd-12 detected by the rotational position detectors 411 and 412 are input to the control unit 61b. The controller 61b determines a correct rotor position among the rotor positions estimated based on the rotor positions θd-11 and θd-12 and the neutral point potential corrected in the same manner as in the first embodiment. Then, the control unit 61b creates a gate command signal to be given to the output pre-driver 313 of the inverter 31 of the system 1 using the rotor position determined to be correct.

In system 2, the rotor positions θd-21 and θd-22 detected by the rotational position detectors 421 and 422 are input to the control unit 62b. The controller 62b determines the correct rotor position among the rotor positions estimated based on the rotor positions θd-21 and θd-22 and the neutral point potential corrected in the same manner as in the first embodiment. Then, the control unit 62b creates a gate command signal to be given to the output pre-driver 323 of the inverter 32 of the system 2 using the rotor position determined to be correct.

FIG. 16 is a block diagram of the control unit 61b of the system 1. In the control unit 61b, so-called vector control is applied as in the first embodiment (FIG. 6). In addition, about the structure of the control part 62b of the system | strain 2, since it is the same as that of the control part 61b, description is abbreviate | omitted.

As FIG. 16 shows, the control part 61b is provided with the detection position determination means 626 in addition to the structure of the control part 61 of Embodiment 1 (FIG. 6).

The detection position determination means 626 detects whether there is an abnormality such as a failure of the rotational position detector 411 and the rotational position detector 412, detects the rotor position θd-21, θd-22, and the neutral point potential detection value with the offset corrected. The determination is made based on the rotor position θd−m estimated by the rotational position estimation unit 623 based on Vn−m ′. Then, the rotor position output from the rotational position detector having no abnormality is output as the rotational position θd-31 used for motor control. When it is determined that both the rotational position detector 411 and the rotational position detector 412 are abnormal, the detection position determination unit 626 uses the rotor position θd−m estimated by the rotational position estimation unit 623 for motor control. The rotation position θd-31 to be used is output.

FIG. 17 is a flowchart showing the determination process executed by the detection position determination means 626 in the system 1. The determination process executed by the detection position determination unit in the system 2 is the same.

First, in step S11, the detection position determination means 626 determines whether θd-11, which is the output of the rotational position detector 411, and θd-12, which is the output of the rotational position detector 412, are substantially the same. For example, if the magnitude of the difference between θd-11 and θd-12 is less than or equal to a preset value, it is determined that they are substantially the same. If θd-11 and θd-12 substantially match (Yes in step S11), the process proceeds to step S12. If θd-11 and θd-12 do not match, the process proceeds to step S13 (No in step S11).

In step S12, the detection position determination means 626 outputs θd-11 as the correct rotor position θd-31. That is, θd-11 is used for motor control in the control unit 61b. In this step S12, the detection position determination means 626 may output θd-12 as θd-31 instead of θd-11.

Here, if θd-11 and θd-12 do not match, it can be determined that either the rotational position detector 411 or the rotational position detector 412 is abnormal. Therefore, in step S13 and step S15, any of the rotational position detectors 411 and 412 is abnormal using the estimated rotor position θdm output from the rotational position estimation unit 623. It is determined whether there is any.

In step S13, the detection position determination unit 626 determines whether θd-11 and θd−m are substantially the same. For example, when the magnitude of the difference between θd-11 and θd−m is less than or equal to a preset value, it is determined that they are substantially the same. When θd-11 and θd−m substantially match (Yes in step S13), it is determined that the rotational position detector 411 is normal, and the process proceeds to step S14, where θd-11 and θd−m do not match. The rotational position detector 411 is determined to be abnormal, and the process proceeds to step S15 (No in step S13).

In step S14, the detection position determination means 626 outputs θd-11 as the correct rotor position θd-31. That is, θd-11 is used for motor control in the control unit 61b.

In step S15, the detection position determination unit 626 determines whether θd-12 and θd-m substantially match. For example, if the magnitude of the difference between θd−12 and θd−m is less than or equal to a preset value, it is determined that they are approximately the same. When θd-12 and θd-m substantially match (Yes in step S15), it is determined that the rotational position detector 412 is normal, and the process proceeds to step S16, where θd-12 and θd-m do not match. The rotational position detector 412 is determined to be abnormal (No in step S15), and the process proceeds to step S17.

In step S16, the detection position determination means 626 outputs θd-12 as the correct rotor position θd-31. That is, θd-12 is used for motor control in the control unit 61b.

In step S17, since it is determined in steps S13 and S15 that both the rotational position detectors 411 and 412 are abnormal, the detected position determining means 626 outputs θd−m as the correct rotor position θd−31. To do. That is, in the control unit 61b, θd−m is used for motor control.

It should be noted that the rotor positions of θd-11, θd-12, and θd-m are preferably positions at the same timing. For example, the three positions can be compared at the same timing by correcting the detection timing of the rotational position detector or correcting each position data by interpolation or the like. Thereby, the determination accuracy of the abnormality of the rotational position detector is improved.

As described above, according to the third embodiment, it is possible to determine which of the plurality of redundantly provided rotational position detectors is abnormal based on the estimated rotor position. As a result, even if any of the plurality of rotational position detectors is abnormal, a normal rotational position detector is selected, and motor control is executed in the same way as when it is normal (when there is no failure). Thus, the desired motor torque can be continuously output. Furthermore, even when the plurality of rotational position detectors are both abnormal, the motor control can be executed using the estimated rotor position, so that the motor drive can be maintained.

Note that the detection position estimation means and the rotational position estimation unit in the third embodiment are functions of a microcomputer constituting the control system, and can be realized without adding hardware. For this reason, according to the third embodiment, the reliability of the motor control device can be improved without increasing the cost of the motor control device.

Similarly to the above-described third embodiment, when a plurality of rotational position detectors and estimation of the rotor position based on the neutral point potential are used in combination, there are the following modifications.

In this modification, after it is determined that a plurality of rotational position detectors (for example, 411 and 412 in FIG. 15) included in the system 1 are abnormal, the control unit (61b) of the system 1 performs the neutral point potential Vn−. Excluding the rotational position detector determined to be abnormal based on the rotor position estimated based on m and the rotor position estimated by the control unit (62b) of system 2 based on the neutral point potential Vn-s It is determined whether there is any abnormality in the remaining rotational position detectors. Thereby, the determination accuracy of the presence or absence of abnormality of the remaining rotational position detectors is improved.
(Embodiment 4)
Hereinafter, Embodiment 4 of the present invention will be described with reference to FIG. Note that differences from the first embodiment will be mainly described.

FIG. 18 is a block diagram illustrating a configuration of the control unit 61c of the system 1 in the motor control apparatus according to the fourth embodiment of the present invention. In addition, the control part of the system | strain 2 also has the same structure. For this reason, illustration and description of the control unit of the system 2 are omitted.

As shown in FIG. 18, in the fourth embodiment, the control unit 61 c includes a medium / high speed position estimator 627 and an estimated position changeover switch 628 in addition to the configuration of the control unit 61 of the first embodiment (FIG. 6).

The medium / high speed position estimator 627 calculates the rotor position θdc2 from the constants (inductance and winding resistance) of the permanent magnet synchronous motor 4 based on the dq axis voltage commands Vd * and Vq * and the dq axis current detection values Id and Iq. Is estimated. This is a known rotor position estimation means based on the induced voltage, and a description of a specific calculation method is omitted. Various means are known as rotor position estimation means based on the induced voltage, and detailed description is omitted, but any means may be applied.

The estimated position changeover switch 628 is configured to output θdc2 output from the medium / high speed position estimator 627 and θd−m estimated based on the neutral point potential detection value Vn−m ′ corrected by the rotational position estimating unit 623. It selects according to speed (rotation speed), and outputs as rotor position (theta) dc3 used for control. That is, the rotor position estimation algorithm is changed according to the motor speed. For example, assuming that a speed equal to or higher than a predetermined value is medium high speed, and a speed smaller than the predetermined value is low speed, the estimated position changeover switch 628 selects θdc2 at medium high speed and θd−m at low speed. In the fourth embodiment, the motor speed ω1 is calculated by the speed calculation means 620 based on θdc3.

Instead of switching between θdc2 and θd−m, weighting is applied to θd−m and θdc2 so that θd−m is dominant in the low speed range and θdc2 is dominant in the medium and high speed range. Thus, the rotor position θdc3 may be calculated. In this case, since the control based on the neutral point potential and the control based on the induced voltage are gradually switched, the stability of the control is improved when switching between the low speed region and the high speed region. Further, hysteresis may be given to the rotation speed for switching between θd−m and θdc2. Thereby, hunting at the time of switching can be prevented.

In the fourth embodiment, θdc2 and θdm are switched according to the motor speed calculated by the speed calculation means 620, but are not limited to this, and are detected by a rotational position sensor (magnetic pole position sensor, steering angle sensor, etc.). Depending on the motor speed, θdc2 and θdm may be switched.

As described above, according to the fourth embodiment, the accuracy of the rotor position used for motor control is improved in a wide speed range from the low speed range to the medium to high speed range. Or, reliability is improved.

In addition, even in a motor structure in which a large harmonic component is generated in the neutral point potential as the speed increases, the rotor position is estimated using the speed-induced voltage only in the high-speed range. Position sensorless drive can be realized.

Note that the medium / high speed position estimator 627 and the estimated position changeover switch 628 in FIG. 18 may be applied to the above-described first to third embodiments.
(Embodiment 5)
Hereinafter, Embodiment 5 of the present invention will be described with reference to FIG. Note that differences from the first embodiment will be mainly described.

FIG. 19 is a block diagram showing a configuration of a three-phase synchronous motor control device (hereinafter, referred to as “motor control device”), which is Embodiment 5 of the present invention.

As shown in FIG. 19, in the fifth embodiment, in addition to the neutral point potential Vn-m of the three-phase winding 41 of the own system, the control unit 61d of the system 1 has three systems of the other system. The neutral point potential Vn-s of the phase winding 42 is input. Similarly, the neutral point potential of the other system (system 1) is input to the control unit 62d of the system 2 in addition to the neutral point potential Vn-m of the own system.

The control units 61d and 62d have the functions of the control units 61 and 62 in the first embodiment and further have the following functions. In the following, the control unit 61d will be described, and the control unit 62d is the same, and thus the description thereof will be omitted.

In the fifth embodiment, the control unit 61d of the system 1 uses the magnetic point between the three-phase winding 41 and the three-phase winding 42 based on the neutral point potential Vn-s of the three-phase winding 42 of the system 2. Rotation using the neutral point potential Vn-m of the three-phase winding 41 when not affected by the magnetic interference, or the neutral point potential Vn-m of the three-phase winding 41 from which the influence of magnetic interference is removed Estimate the child position.

First, the neutral point potential Vn-m of the three-phase winding 41 when not affected by magnetic interference will be described.

When a voltage is applied to the three-phase winding 42, the neutral point potential Vn-m of the three-phase winding 41 varies due to magnetic interference. Therefore, the neutral point potential Vn-m of the three-phase winding 41 detected when no voltage is applied to the three-phase winding 42 is not affected by magnetic interference. When no voltage is applied to the three-phase winding 42, the voltage vector output from the inverter 32 is one of zero vectors, that is, V (0,0,0) and V (1,1,1).

In the case of V (0, 0, 0), the upper semiconductor switching elements (Sup2, Svp2, Swp2) of the inverter main circuit 321 are OFF and the lower semiconductor switching elements (Sun2, Svn2, Swn2) are ON. The neutral point potential Vn-s of the three-phase winding 42 is a potential on the low potential side of the DC power supply 5, which is the ground potential (hereinafter referred to as zero) in the fifth embodiment.

In the case of V (1, 1, 1), the upper semiconductor switching elements (Sup2, Svp2, Swp2) of the inverter main circuit 321 are ON, and the lower semiconductor switching elements (Sun2, Svn2, Swn2) are OFF. Therefore, the neutral point potential Vn-s of the three-phase winding 42 becomes a high potential side potential (hereinafter referred to as E) of the DC power supply 5.

Therefore, the control unit 61 determines whether the neutral point potential Vn-s of the three-phase winding 42 is zero or E. That is, the control unit 61 determines whether or not a voltage is applied to the three-phase winding 42 based on the neutral point potential Vn−s of the three-phase winding 42. When the control unit 61 determines that the neutral point potential Vn-s of the three-phase winding 42 is zero or E, that is, determines that no voltage is applied to the three-phase winding 42, the control unit 61 detects that. Based on the neutral point potential Vn−m of the three-phase winding 41, the rotor position is estimated as in the first embodiment.

The control unit 61d in the system 1 and the control unit 62d in the system 2 have a configuration in which the phase of the triangular wave carrier for PWM is shifted by a predetermined amount, so that the other system is V (0, 0, 0) or V ( The neutral point potential in the own system can be reliably detected at the timing of (1, 1, 1). Preferably, by shifting the phase by 90 degrees, the neutral point potential in the own system can be reliably detected at the timing when the other system becomes the zero vector.

Next, the neutral point potential Vn-m of the winding 41 from which the influence of magnetic interference between the winding 41 and the three-phase winding 42 is removed will be described.

For each voltage vector output from the system 2, the fluctuation of the neutral point potential Vn-m of the winding 41 due to the influence of magnetic interference between the winding 41 and the three-phase winding 42 is measured in advance, and the measured value And data representing the relationship between the voltage vector and the voltage vector (for example, table data) are created and set in the control unit 61d.

Based on the neutral point potential Vn-s of the three-phase winding 42, the control unit 61d determines that the voltage vector output from the inverter 32 to the three-phase winding 42 is V (0, 0, 0), V (1, 1, 1), V (1, 0, 0), V (0, 1, 0), V (0, 0, 1), V (1, 1, 0), V (1, 0, 1) Determine if there is. The controller 61d reads the value of the fluctuation of the neutral point potential Vn-m corresponding to the determined voltage vector from the above data, and corrects the neutral point potential Vn-m using this value. Then, the control unit 61d estimates the rotor position based on the corrected neutral point potential as in the first embodiment.

As described above, according to the fifth embodiment, the control unit estimates the rotational position based on the neutral point potential of the three-phase winding of the own system and the neutral point potential of the three-phase winding of the other system. In the rotor position estimation based on the neutral point potential of the own system, it is possible to eliminate the influence of magnetic interference caused by voltage application by the inverter of another system. Thereby, the estimation accuracy of the rotor position is improved. For this reason, in a motor drive system in which one permanent magnet synchronous motor is driven by two inverters, position sensorless drive at an extremely low speed is possible.
(Embodiment 6)
FIG. 20 shows a configuration of an electric power steering apparatus that is Embodiment 6 of the present invention.

As shown in FIG. 20, in the electric power steering apparatus 8, the rotational torque of the steering wheel 81 is detected by a torque sensor 82, and the inverters 31 (system 1) and 32 in the motor control apparatus 3 are detected according to the detected rotational torque. (System 2) drives and controls the permanent magnet synchronous motor (three-phase winding 41 (system 1), three-phase winding 42 (system 2)). As a result, the motor torque generated by the permanent magnet synchronous motor is transmitted to the steering mechanism 84 via the steering assist mechanism 83. Thus, when the steering wheel 81 is operated by the driver, the tire 85 is steered by the steering mechanism 84 while the electric power steering device 8 assists the steering force in accordance with the operation input to the steering wheel 81.

The motor control device of the third embodiment (FIGS. 15 and 16) is applied to the motor control device 3 of the sixth embodiment. Accordingly, one permanent magnet synchronous motor is driven by the two inverters 31 and 32. Inverters 31 and 32 are controlled based on a rotor position detected by a plurality of redundantly provided rotational position detectors and a rotor position estimated based on a neutral point potential.

According to the sixth embodiment, as in the third embodiment, it is possible to determine which of the plurality of rotational position detectors is out of order based on the estimated rotor position. Even if one of them fails, a normal rotational position detector can be selected and motor control can be executed in the same way as normal (when there is no failure) to continue outputting the desired motor torque. it can. For this reason, the electric power steering apparatus can continue the assist operation normally.

Furthermore, even if a plurality of rotational position detectors fail together, the motor control can be continued using the estimated rotor position, so that the electric power steering device can continue the assist operation. For example, even when a vehicle tire rides on a step, the electric power steering device can continuously assist the steering force.

Also, when a plurality of rotational position detectors fail, the motor control can be continued using the estimated rotor position. Thereby, while notifying a driver | operator that it is a failure, the output of a permanent magnet motor can be reduced gradually and it can prevent falling into an assist stop suddenly. As a result, when both of the plurality of rotational position detectors included in the electric power steering apparatus fail or an abnormality occurs, the driver can safely stop the vehicle.

Also, the rotational position estimating means can be realized without adding hardware. For this reason, according to the sixth embodiment, the reliability of the electric power steering apparatus can be improved without increasing the cost.

In the sixth embodiment, the motor control device 3 is not limited to the third embodiment, and the first, second, and fourth embodiments may be applied.

In addition, this invention is not limited to embodiment mentioned above, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

For example, the number of inverters that drive one permanent magnet synchronous motor is not limited to two, and any number of inverters may be used. The three-phase synchronous motor is not limited to a permanent magnet synchronous motor, but may be a wound field synchronous motor. Moreover, you may use the detected value of the output voltage of an inverter, or a motor terminal voltage as information which shows the drive state of the inverter of another system used for estimation of the rotor position of an own system.

DESCRIPTION OF SYMBOLS 3 ... Motor control apparatus, 4 ... Permanent magnet synchronous motor, 5 ... DC power supply, 8 ... Electric power steering apparatus, 31, 32 ... Inverter, 41, 41a ... Three-phase winding, 42, 42a ... Three-phase winding, 61 , 61a, 61b, 61c, 61d ... control unit, 62, 62a, 62b, 62d ... control unit, 63 ... control unit communication unit, 81 ... steering wheel, 82 ... torque sensor, 83 ... steering assist mechanism, 84 ... steering mechanism , 85, tires, 311, inverter main circuit, 312, one-shunt current detector, 313, output pre-driver, 321, inverter main circuit, 322, one-shunt current detector, 323, output pre-driver, 411, 412, 421 , 422... Rotational position detector, 611... Q-axis current command generating means, 612. ... subtraction means, 613b ... subtraction means, 614a ... d-axis current control means, 614b ... q-axis current control means, 615 ... dq reverse conversion means, 616 ... PWM generation means, 617 ... current reproduction means, 618 ... dq conversion means, 619 ... Sample / hold means, 620 ... Speed calculation means, 621 ... Pulse shift means, 622 ... Neutral point potential detection part, 623 ... Rotation position estimation part, 624 ... Neutral point potential offset calculation means, 625 ... Potential detection means , 626 ... detection position determination means, 627 ... medium / high speed position estimator, 628 ... estimated position changeover switch

Claims (25)

A three-phase synchronous motor comprising a first three-phase winding and a second three-phase winding;
A first inverter connected to the first three-phase winding;
A second inverter connected to the second three-phase winding;
A first rotor that estimates a rotor position of the three-phase synchronous motor based on a neutral point potential of the first three-phase winding, and controls the first inverter based on the estimated rotor position. A control unit;
A second position that estimates the rotor position of the three-phase synchronous motor based on a neutral point potential of the second three-phase winding, and controls the second inverter based on the estimated rotor position; A control unit of
In a control device for a three-phase synchronous motor comprising:
The control device for a three-phase synchronous motor, wherein the first control unit corrects a neutral point potential of the first three-phase winding based on a DC voltage of the first inverter. In the control apparatus of the three-phase synchronous motor according to claim 1,
The control device for a three-phase synchronous motor, wherein the second control unit corrects a neutral point potential of the second three-phase winding based on a DC voltage of the second inverter. In the control apparatus of the three-phase synchronous motor according to claim 1,
The control device for a three-phase synchronous motor, wherein the first control unit corrects a neutral point potential of the first three-phase winding based on a mutual inductance of the first three-phase winding. In the control apparatus of the three-phase synchronous motor according to claim 3,
The first control unit determines a neutral point potential of the first three-phase winding based on a DC voltage of the first inverter and a self-inductance and a mutual inductance of the first three-phase winding. A control device for a three-phase synchronous motor, wherein correction is performed. In the control device for a three-phase synchronous motor according to claim 4,
The control device for a three-phase synchronous motor, wherein the self-inductance and the mutual inductance of the first three-phase winding are changed according to the current of the three-phase synchronous motor. In the control apparatus of the three-phase synchronous motor according to claim 3,
The control device for a three-phase synchronous motor, wherein the first three-phase winding and the second three-phase winding are provided in one stator of the three-phase synchronous motor. In the control apparatus of the three-phase synchronous motor according to claim 6,
A control device for a three-phase synchronous motor, wherein each region in which the first three-phase winding and the second three-phase winding are provided in the one stator is divided. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit is configured to control the first inverter based on a DC voltage of the first inverter, a self-inductance and a mutual inductance of the first three-phase winding, and a self-inductance and a mutual inductance of the second three-phase winding. A control device for a three-phase synchronous motor, wherein the neutral point potential of one three-phase winding is corrected. In the control apparatus of the three-phase synchronous motor according to claim 1,
The three-phase synchronous motor, wherein the first control unit corrects a neutral point potential of the first three-phase winding so as to remove an offset voltage proportional to a DC voltage of the first inverter. Control device. In the control device for a three-phase synchronous motor according to claim 9,
The first control unit corrects a neutral point potential of the first three-phase winding based on a DC voltage of the first inverter that has been subjected to filter processing or average value processing. A control device for a three-phase synchronous motor. In the control apparatus of the three-phase synchronous motor according to claim 1,
A control device for a three-phase synchronous motor, wherein a neutral point potential of the first three-phase winding is corrected using an offset voltage value corresponding to a switching pattern in the first inverter. In the control apparatus of the three-phase synchronous motor according to claim 1,
The control device for a three-phase synchronous motor, wherein the first control unit constantly corrects a neutral point potential of the first three-phase winding. In the control apparatus of the three-phase synchronous motor according to claim 1,
The three-phase synchronous motor is driven by a plurality of inverters including the first inverter and the second inverter. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit estimates a rotor position of the three-phase synchronous motor based on a neutral point potential of the first three-phase winding and a neutral point potential of the second three-phase winding. A control device for a three-phase synchronous motor. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit estimates a rotor position of the three-phase synchronous motor based on the corrected neutral point potential of the first three-phase winding. . The control device for a three-phase synchronous motor according to claim 15,
The first control unit receives a rotor position of the three-phase synchronous motor detected by a rotational position detector,
The first control unit is estimated based on a rotor position of the three-phase synchronous motor detected by the rotational position detector and a corrected neutral point potential of the first three-phase winding. A control device for a three-phase synchronous motor, wherein the presence / absence of abnormality of the rotational position detector is determined by comparing the rotor position of the three-phase synchronous motor. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit estimates a rotor position of the three-phase synchronous motor based on a neutral point potential of the first three-phase winding when a rotation speed of the three-phase synchronous motor is smaller than a predetermined value. And driving the three-phase synchronous motor. The control device for a three-phase synchronous motor according to claim 17,
The control apparatus for a three-phase synchronous motor, wherein the rotational speed is detected by a steering angle sensor. The control device for a three-phase synchronous motor according to claim 17,
When the rotational speed is greater than or equal to the predetermined value, the first control unit estimates the rotor position of the three-phase synchronous motor based on the induced voltage of the first three-phase winding, and the three-phase synchronous A control device for a three-phase synchronous motor, wherein the motor is driven. The control device for a three-phase synchronous motor according to claim 17,
The first control unit has a hysteresis in the rotational speed when estimating a rotor position of the three-phase synchronous motor based on a neutral point potential of the first three-phase winding. Control device for three-phase synchronous motor. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit detects a neutral point potential of the first three-phase winding, estimates a rotor position, and drives the three-phase synchronous motor. Control device. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit receives a rotor position of the three-phase synchronous motor detected by each of the first rotational position detector and the second rotational position detector,
The first control unit includes a rotor position of the three-phase synchronous motor detected by each of the first rotational position detector and the second rotational position detector, and the first three-phase winding. Determining whether there is an abnormality in the first rotational position detector and the second rotational position detector based on a rotor position of the three-phase synchronous motor estimated based on a neutral point potential of A control device for a three-phase synchronous motor. The control device for a three-phase synchronous motor according to claim 22,
When the first controller determines that both the first rotational position detector and the second rotational position detector are abnormal, the first controller estimates based on a neutral point potential of the first three-phase winding. A control device for a three-phase synchronous motor, wherein the three-phase synchronous motor is driven based on a rotor position of the three-phase synchronous motor. The control device for a three-phase synchronous motor according to claim 22,
A plurality of rotational position detectors including the first rotational position detector and the second rotational position detector;
After the first control unit determines that both the first rotational position detector and the second rotational position detector are abnormal, the first rotational position detection of the plurality of rotational position detectors. The rotor position of the three-phase synchronous motor estimated based on the neutral point potential of the first three-phase winding, the presence or absence of the remaining abnormality except for the generator and the second rotational position detector; 3. A control device for a three-phase synchronous motor, characterized in that a determination is made based on a rotor position of the three-phase synchronous motor estimated based on a neutral point potential of two three-phase windings. In the control apparatus of the three-phase synchronous motor according to claim 1,
The first control unit receives a rotor position of the three-phase synchronous motor detected by each of the first rotational position detector and the second rotational position detector,
The first control unit includes a rotor position of the three-phase synchronous motor detected by each of the first rotational position detector and the second rotational position detector, and the first three-phase winding. Determining the presence or absence of an abnormality in the first rotational position detector and the second rotational position detector based on the rotor position of the three-phase synchronous motor estimated based on the neutral point potential;
The first controller is
The three-phase synchronous motor detected by the first rotational position detector or the second rotational position detector when the first rotational position detector or the second rotational position detector determines that there is no abnormality. Driving the three-phase synchronous motor based on the rotor position of
The rotor of the three-phase synchronous motor estimated based on the neutral point potential of the first three-phase winding when both the first rotational position detector and the second rotational position detector are determined to be abnormal A control device for a three-phase synchronous motor, wherein the three-phase synchronous motor is driven based on a position.

Images