JP2010148270A - Controller for permanent magnet type rotating electrical machine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a controller for permanent magnet type rotating electrical machine, which derives the amount of correction of an error dependent on the number of revolutions included in the detected position of a magnetic pole, an error in attachment, an error due to secular change, and an error due to dispersion of products. <P>SOLUTION: A controller for permanent magnet type rotating electrical machine executes the processing of dq vector control by setting the amount of positional correction of a magnetic pole to a predetermined temporary set value, while keeping both a d-axis current command and a q-axis current command in dq vector control zero on a dq coordinate system, where the direction of the field of a rotor is d axis and the direction orthogonal to the d axis is q axis, in each state where the rotor is rotating in two or more roughly constant numbers of revolutions. It computes an error, based on a predetermined operational expression where the d-axis voltage command and the q-axis voltage command are parameters, from the d-axis voltage command and the q-axis voltage command obtained at execution of dq vector control, and derives number of revolutions-error properties, based on the error in each state and the number of revolutions at that time. Furthermore, based on the number of revolutions-error properties, it gets the amount of positional correction of the magnetic pole geared to the number of revolutions of the rotor for correcting the detected position of the magnetic pole. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a control device for a permanent magnet type rotating electrical machine.

  In a permanent magnet type rotating electric machine (for example, a three-phase DC brushless motor) in which a rotor (rotor) and a stator (stator) are each provided with a permanent magnet and an armature, the rotor is controlled in order to control the generated torque and rotation speed. Of the armature applied voltage (specifically, the applied voltage of the winding of each phase constituting the armature, hereinafter referred to as the armature applied voltage) in accordance with the magnetic pole position (more specifically, the rotational angle position of the magnetic pole) Need to operate. For this reason, this type of rotating electric machine is provided with a magnetic pole position detector that detects the magnetic pole position of the rotor, and the phase of the armature applied voltage is manipulated according to the detected magnetic pole position. The magnetic pole position detector is a Hall element, an encoder, a resolver, or the like.

  In the control of the permanent magnet type rotating electrical machine equipped with the magnetic pole position detector as described above, the magnetic pole position is caused by the alignment at the time of mounting the magnetic pole position detector or the manufacturing accuracy of the magnetic pole position detector itself. The magnetic pole position detected by the detector often causes an error with respect to the actual magnetic pole position. If there is such an error, operating the phase of the armature voltage using the detected magnetic pole position as it is will cause a reduction in the power factor and efficiency of the rotating electrical machine.

  For this reason, for example, as can be seen in Patent Document 1, a technique for correcting the detected value of the magnetic pole position is known. In the technique described in Patent Document 1, in a rotating electrical machine (cylindrical machine) in which the rotor magnet is cylindrical, when the generated torque T of the rotating electrical machine is proportional to the q-axis current Iq and the load torque is constant, When the armature voltage is manipulated so that the armature current (phase current) is minimized, the ratio between the d-axis current command value and the armature current, or the ratio between the d-axis current command value and the q-axis current command value is It is noted that there is a certain correlation between the error angle between the magnetic pole position detected by the magnetic pole position detector and the actual magnetic pole position. In the technique of the publication, an error angle is calculated based on the value of the ratio, and the detected value of the magnetic pole position is corrected based on the calculated error angle to control the rotating electrical machine.

  However, since the technique described in Patent Document 1 is based on the premise that the torque T generated by the rotating electrical machine is proportional to the q-axis current Iq, the rotating electrical machine (the salient pole) in which the rotor magnet is a salient pole type. It cannot be applied to the machine. That is, the torque T generated by the permanent magnet type rotating electrical machine is expressed by the following equation (A) as described in Patent Document 1.

T = Φ · Iq + (Ld−Lq) · Id · Iq (A)
Where Φ: magnetic flux, Ld, Lq: d, q-axis inductance,
Id, Iq: d, q axis current

  In this case, in a cylindrical machine having a cylindrical magnet, since Ld = Lq, the torque T is proportional to the q-axis current Iq. However, in a salient pole machine in which the magnet has a salient pole shape, since Ld ≠ Lq, the torque T is not proportional to the q-axis current Iq. For this reason, in the salient pole machine, the precondition of the technique of Patent Document 1 is not satisfied, and the detection value of the magnetic pole position cannot be corrected appropriately.

The technique described in Patent Document 2 solves the above problem. FIG. 12 is a block diagram showing an internal configuration of a control device for a permanent magnet type rotating electrical machine disclosed in Patent Document 2 in which a rotor is connected to an output shaft of an engine. FIG. 13 is a graph showing changes in engine speed, d-axis current, q-axis current, and voltage phase when the control device shown in FIG. 12 calculates the magnetic pole position correction amount. The voltage phase is an arc tangent (tan ⁻¹ (Vdc / Vqc)) of the ratio between the d-axis voltage command value Vdc and the q-axis voltage command value Vqc.

  As shown in FIG. 13, the control device shown in FIG. 12 is rotated while the rotor 4 of the permanent magnet type rotating electrical machine 1 is rotated at a constant rotational speed by the engine 3 and the armature current is substantially zero. A dq vector control process for handling the rotating electrical machine 1 is executed in a dq coordinate system in which the field direction of the child 4 is the d-axis and the direction orthogonal to the d-axis is the q-axis, and d obtained by this dq vector control process A magnetic pole position correction amount for correcting the magnetic pole position detected by the magnetic pole position detector 8 is obtained so that the shaft voltage command value becomes substantially zero. The control device manipulates the phase of the armature voltage using the magnetic pole position obtained by correcting the magnetic pole detection position with the magnetic pole position correction amount.

  Hereinafter, derivation of the magnetic pole position correction amount based on the technique of Patent Document 2 will be described. The dq coordinates in FIG. 14A are dq coordinates (hereinafter referred to as real coordinates dq) with the actual field direction of the rotor 4 as the d axis, and the dc-qc coordinates in FIG. Indicates a dq coordinate determined by the magnetic pole position detected by the magnetic pole position detector 8 (hereinafter referred to as a magnetic pole detection position) (dq coordinate in the above-described dq vector control process; hereinafter referred to as a command axis coordinate dc-qc). Yes.

  Here, the rotor 4 of the permanent magnet type rotating electrical machine 1 is rotating, and the armature current I (current flowing through each phase of the armature) of the rotating electrical machine 1 is “0” (hereinafter referred to as “the armature current I”). This state is referred to as a zero current state). In this zero current state, the armature applied voltage V (applied voltage for each phase of the armature) is equal to the counter electromotive voltage E generated by the field of the rotor 4. In this case, it is assumed that there is no error in the magnetic pole detection position with respect to the true magnetic pole position. That is, as shown in FIG. 14A, it is assumed that the actual coordinate dq matches the command axis coordinate dc-qc. At this time, the d-axis voltage command value Vdc (voltage command value on the command axis dc) obtained by the dq vector control process is Vdc = 0, and the q-axis voltage command value Vqc is Vqc = E.

  Therefore, in a state where the d-axis voltage command value Vdc obtained by the dq vector control process in the zero current state is “0”, the magnetic pole position is correctly detected. This means that in order to correctly grasp the magnetic pole position, it is only necessary to correct the magnetic pole detection position so that the d-axis voltage command value Vdc is “0” in the zero current state.

Further, it is assumed that there is an error in the magnetic pole detection position with respect to the true magnetic pole position in the zero current state. For example, as shown in FIG. 14B, it is assumed that the command axis coordinate dc-qc has an error of the angle θofs with respect to the actual coordinate dq (hereinafter, the angle θofs is referred to as a magnetic pole position error angle θofs). At this time, the d-axis voltage command value Vdc (voltage command value on the command axis dc) obtained by the dq vector control process is Vdc ≠ 0, and the q-axis voltage command value Vqc (voltage command value on the command axis qc) is Vqc ≠ E. The square root √ (Vdc ² + Vqc ² ) of the sum of the square value of Vdc and the square value of Vqc is equal to the magnitude of the back electromotive force E. Further, in this case, the ratio (Vdc / Vqc) between the d-axis voltage command value Vdc and the q-axis voltage command value Vqc is equal to the tangent tan θofs of the magnetic pole position error angle θofs. That is, the following expression (1) is established.

θofs = tan ^-1 (Vdc / Vqc) (1)

  The magnitude of the back electromotive force E, and hence the magnitude of the d-axis voltage command value Vdc and the q-axis voltage command value Vqc itself, depends on the rotational speed of the rotor, but the equation (1) is If there is, it is established regardless of the rotational speed of the rotor 4. Moreover, Formula (1) is equivalent to following Formula (2) or (3).

θofs = sin ⁻¹ {Vdc / (√ (Vdc ² + Vqc ² ))} (2)

θofs = cos ⁻¹ {Vqc / (√ (Vdc ² + Vqc ² ))} (3)

  Accordingly, from the d-axis voltage command value Vdc and the q-axis voltage command value Vqc obtained by the dq vector control process in the zero current state, the magnetic pole position error angle θofs is calculated by the equation (1), (2) or (3). It can be obtained. If the magnetic pole detection position is corrected by this magnetic pole position error angle θofs, the correct magnetic pole position can be grasped. For example, if the rotation angle position of the magnetic pole corresponding to the magnetic pole detection position is θact, the angle θact−θofs obtained by subtracting the magnetic pole position error angle θofs from the rotation angle position θact represents the correct magnetic pole position.

  The principle described above is a matter that is established regardless of whether the magnet of the rotor 4 is cylindrical or salient.

  As described above, in the technique described in Patent Document 2, a predetermined arithmetic expression (formula (1) or (2)) is obtained from the d-axis voltage command value Vdc and the q-axis voltage command value Vqc obtained by the dq vector control process in the zero current state. Since the magnetic pole position correction amount θofs is obtained based on (2) or (3)), the magnetic pole position obtained by correcting the magnetic pole detection position with the magnetic pole position correction amount θofs is whether the permanent magnet type rotating electrical machine 1 is a cylindrical machine. Regardless of whether it is a salient pole machine, it matches the actual magnetic pole position of the rotor 4. Therefore, by operating the phase of the armature voltage according to the corrected magnetic pole position, the operation control (torque control and speed control) of the rotating electrical machine 1 can be performed without impairing the efficiency and power factor of the rotating electrical machine 1. It can be done.

JP 2001-8486 A (paragraph numbers 0008, 0018 to 0021) Japanese Patent No. 3688673

  Each component of the control device shown in FIG. 12 and a signal input IF obtained from the magnetic pole position detector 8 have phase characteristics. For this reason, the magnetic pole position detected by the magnetic pole position detector 8 includes an error corresponding to the rotational speed of the rotor 4 of the permanent magnet type rotating electrical machine 1. In the technique of Patent Document 2 described above, the magnetic pole position correction amount is obtained when the rotor 4 of the permanent magnet type rotating electrical machine 1 is rotated at a constant rotational speed by the engine 3. This is a value for correcting the magnetic pole detection position including an error corresponding to the rotational speed when obtaining the position correction amount. Therefore, even if the magnetic pole position detected at a rotational speed different from the constant rotational speed is corrected with the magnetic pole position correction amount, an error depending on the rotational speed remains included.

  In order to solve this problem, it is conceivable to store an error depending on the rotational speed in a memory for each rotational speed and correct the magnetic pole detection position according to the rotational speed. However, with this method, errors due to alignment when mounting the magnetic pole position detector (hereinafter referred to as mounting errors), errors due to aging of components for detecting or correcting the magnetic pole position of the rotor, Errors due to variations cannot be corrected. In order to solve the above problem, there is also a method of using a part with very little secular change and product variation. However, such components are generally expensive, and this method is not preferable because the mounting error cannot be corrected.

  An object of the present invention is to provide a control device for a permanent magnet type rotating electrical machine capable of deriving a correction amount of an error depending on the number of rotations included in a magnetic pole detection position, an installation error, an error due to secular change, and an error due to product variation. That is.

  In order to solve the above problems and achieve the object, a control device for a permanent magnet type rotating electrical machine according to a first aspect of the present invention includes a rotor (for example, the rotor 4 in the embodiment) and a stator ( For example, the permanent magnet type rotating electric machine in which the stator 6 in the embodiment is provided with a permanent magnet (for example, the permanent magnet 5 in the embodiment) and an armature (for example, the armature 7 in the embodiment). A magnetic pole position detector (for example, a magnetic pole position detector 8 in the embodiment) for detecting the magnetic pole position of the rotor in the motor 1 (for example, the motor 1 in the embodiment), and the rotor at a substantially constant rotational speed. An error calculator (for example, the phase corrector 101 in the embodiment) that calculates an error of the magnetic pole position detected by the magnetic pole position detector in the rotating state with respect to the actual magnetic pole position is substantially different from the rotor. For each state that is rotating at a constant rotational speed, Rotation indicating the relationship between the rotor rotation speed and the magnetic pole position error based on a plurality of errors calculated by the difference calculation section and the rotation speed of the rotor when the error calculation section calculates each error. A characteristic deriving unit (for example, the phase corrector 101 in the embodiment) for deriving a number-error characteristic, and a magnetic pole position detected by the magnetic pole position detecting unit based on the rotation speed-error characteristic A correction amount determination unit (for example, the phase corrector 101 in the embodiment) for obtaining a magnetic pole position correction amount according to the number of rotations of the rotor, and the magnetic pole position detected by the magnetic pole position detection unit determine the correction amount. A current control unit for controlling the armature current flowing in the armature while manipulating the phase of the armature voltage applied to the armature according to the magnetic pole position corrected by the magnetic pole position correction amount obtained by the unit ( For example, the current in the embodiment In the control device for a permanent magnet type rotating electrical machine comprising the command generator 9, the current command switching device 10, the voltage command generator 15, the voltage coordinate converter 16 and the PWM inverter circuit 17), the error calculation section is 2 In each state where the rotor is rotating at two or more different substantially constant rotation speeds, the rotation is performed in a dq coordinate system in which the field direction of the rotor is the d axis and the direction orthogonal to the d axis is the q axis. While maintaining both the d-axis current command value and the q-axis current command value in dq vector control for handling an electric machine at zero, the magnetic pole position correction amount is set to a predetermined temporary setting value, and the dq vector control processing is executed. Then, an error is calculated from the d-axis voltage command value and the q-axis voltage command value obtained during the execution of the dq vector control based on a predetermined arithmetic expression using the d-axis voltage command value and the q-axis voltage command value as variables. It is characterized by that.

  Furthermore, in the control device for the permanent magnet type rotating electrical machine according to the second aspect of the present invention, the error calculation unit calculates the error of the magnetic pole position in each state where the rotor is rotating at two different substantially constant rotational speeds. When calculated, the characteristic deriving unit derives the rotation speed-error characteristic by linear approximation.

  Furthermore, in the control device for a permanent magnet type rotating electrical machine according to the third aspect of the present invention, the error calculation unit determines the position of the magnetic pole in each state where the rotor rotates at three or more different substantially constant rotational speeds. When the error is calculated, the characteristic deriving unit derives the rotation speed-error characteristic by curve approximation.

  According to the control device for the permanent magnet type rotating electrical machine of the inventions according to claims 1 to 3, an error depending on the number of rotations included in the magnetic pole detection position, an installation error of the magnetic pole position detection unit, an error due to secular change, and a product The amount of error correction due to variation can be derived.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  An embodiment of the present invention will be described with reference to FIG. 14 and FIGS. FIG. 1 is a block diagram showing the apparatus configuration of this embodiment. In FIG. 1, the same reference numerals are assigned to components common to FIG. 12. A motor control device 100 that performs current control of a permanent magnet type rotating electrical machine 1 (hereinafter referred to as a motor 1) is provided as a control device for the permanent magnet type rotating electrical machine of the present embodiment.

  In the present embodiment, the motor 1 is mounted on a parallel hybrid vehicle as a prime mover that generates auxiliary output (auxiliary vehicle propulsive force) that assists the output (vehicle propulsive force) of the engine 3 as necessary. It is. A rotor (rotor) 4 of the motor 1 is connected to the output shaft 3 a so as to be able to rotate in conjunction with the output shaft 3 a of the engine 3 and outputs (torque) to be generated by the rotor 4. Along with the output of the engine 3, it is transmitted to the drive wheels of the vehicle via a power transmission device such as a transmission (not shown).

  The motor 1 is a three-phase DC brushless motor in the present embodiment, and its main configuration is schematically shown in the cross-sectional view of FIG. As shown in the figure, the rotor (rotor) 4 of the motor 1 is provided with a permanent magnet 5, and the stator (stator) 6 is composed of three-phase (U-phase, V-phase, W-phase) windings. An armature 7 is provided. The permanent magnet 5 is cylindrical in this embodiment, but may be salient. FIG. 2 also shows dq coordinates with the field direction of the permanent magnet 5 of the rotor 4 as the q-axis and the direction orthogonal thereto as the d-axis.

  Returning to FIG. 1, the magnetic pole position detector 8 that detects the magnetic pole position of the rotor 4 is attached to the motor 1. The magnetic pole position detector 8 is a Hall element, an encoder, a resolver, or the like, and outputs a signal indicating a detected value of the rotation angle θact (rotation angle of the q axis) of the magnetic pole from a predetermined reference rotation position of the rotor 4 to the magnetic pole position. Is output as a detection signal. Note that the detected value of the rotation angle θact obtained by the magnetic pole position detector 8 (hereinafter referred to as the magnetic pole detection angle θact) is generally due to an installation error of the magnetic pole position detector 8 or the like. An error occurs with respect to the magnetic pole position (the actual rotation angle of the magnetic pole).

  The motor control device 100 controls the operation of the motor 1 by the above-described dq vector control, and the d-axis current command values Idc and q in accordance with the torque command value Trc that is the command value of the torque generated by the motor 1. A current command generator 9 for determining an axis current command value Iqc, a set (Idc, Iqc) of a d-axis current command value Idc and a q-axis current command value Iqc output from the current command generator 9 and a d of a value “0” A current command switching unit 10 that selectively outputs a set (0, 0) of the shaft current command value Idc and the q-axis current command value Iqc is provided. In this case, the torque command value Trc input to the current command generator 9 is set according to the driving state (accelerator operation amount, etc.) of the vehicle by an arithmetic processing unit (not shown). Then, the current command generator 9 obtains a d-axis current and a q-axis current required for causing the motor 1 to generate a torque of the input torque command value Trc, and obtains them as a d-axis current command value Idc and a q-axis current command. Output as value Iqc. Further, the current command switching unit 10 switches its output in accordance with a command from a phase corrector 101 described later.

  The motor control device 100 also includes armature current detectors 11u and 11v that detect armature currents Iu and Iv flowing in the U-phase and V-phase of the armature 7 of the motor 1, respectively, and the armature currents Iu and Iv. A current coordinate converter 12 is provided for calculating the d-axis current Id and the q-axis current Iq at the command axis coordinates dc-dq by converting the detected values into coordinates. Here, since the armature 7 has three phases, the current flowing through any one phase is uniquely determined by the current flowing through the other two phases. For example, the current flowing through the W phase is − (Iu + Iv). For this reason, in this embodiment, the armature current detectors 11u and 11v detect currents in two phases (U phase and V phase in this embodiment). Further, the coordinate conversion by the current coordinate converter 12 is performed by the following equation (4) using the magnetic pole rotation angle θ obtained by a subtraction processor 19 described later as indicating the rotation angle of the magnetic pole of the rotor 4.

  Id and Iq obtained by this coordinate conversion are the detected value of the d-axis current at the command axis coordinate dc-qc determined by the magnetic pole rotation angle θ (the dq coordinate determined by θ being the rotational position of the d-axis), the detection of the q-axis current. It has meaning as a value. In the following description, Id and Iq are referred to as d-axis detection current Id and q-axis detection current Iq.

  Further, the motor control device 100 outputs the deviation between the d-axis current command value Idc output from the current command switch 10 and the d-axis detection current Id obtained by the current coordinate converter 12 and the current command switch 10. Subtraction processors 13 and 14 for obtaining deviations between the q-axis current command value Iqc and the q-axis detection current Iq obtained by the current coordinate converter 12, and the deviations (Idc−Id) and (Iqc−Iq). Accordingly, a voltage command generator 15 for obtaining a d-axis voltage command value Vdc and a q-axis voltage command value Vqc, which are command values of applied voltages in the respective axis directions at the command axis coordinates dc-qc, and the d-axis voltage command value Vdc. And voltage coordinate conversion for calculating command values Vuc, Vvc, Vwc (hereinafter referred to as phase voltage command values Vuc, Vvc, Vwc) of the applied voltage of each phase of the armature 7 by coordinate conversion of the q-axis voltage command value Vqc. 16 and the phase voltage command value Vu and a PWM inverter circuit 17 for operating the actual applied voltage of each phase of the armature 7 according to c, Vvc, and Vwc.

  The voltage command generator 15 basically has a d-axis voltage command value based on a feedback control law such as a PI control law so that the deviations (Idc−Id) and (Iqc−Iq) are set to “0”. Vdc and q-axis voltage command value Vqc are obtained. Since this type of voltage command generator 15 is a well-known one, a detailed description thereof will be omitted. However, in addition to the feedback control law processing, the voltage command generator 15 is provided with d and q axes. The d-axis voltage command value Vdc and the q-axis voltage command value Vqc are obtained by performing non-interference control processing for compensating for the interference of the speed electromotive force between them.

  The coordinate conversion by the voltage coordinate converter 16 is performed by the following equation (5) using the same magnetic pole rotation angle θ as that used for the coordinate conversion of the current coordinate converter 12.

  The phase voltage command values Vuc, Vvc, Vwc obtained by this coordinate conversion define the magnitude and phase of the applied voltage of each phase U, V, W of the armature 7, and the PWM inverter circuit 17 The magnitude (amplitude) and the phase of the applied voltage of each phase U, V, W of the armature 7 are manipulated according to the phase voltage command values Vuc, Vvc, Vwc.

  The motor control device 100 further obtains and outputs the magnetic pole position error angle θofs representing the error angle of the magnetic pole detection angle θact by the magnetic pole position detector 8 from the actual rotation angle of the magnetic pole as a magnetic pole position correction amount in the present invention. By subtracting the magnetic pole position error angle θofs (hereinafter referred to as magnetic pole correction angle θofs) output from the phase corrector 101 from the phase corrector 101 from the magnetic pole detection angle θact by the magnetic pole position detector 8, a current coordinate converter 12 and the subtraction processor 19 for determining the magnetic pole rotation angle θ (= θact−θofs) used in the coordinate conversion of the voltage coordinate converter 16 and the rotational speed of the rotor 4 of the motor 1 (details) by differentiating the magnetic pole detection angle θact. Is provided with a speed calculator 20 for obtaining a rotational angular speed) ω = dθact / dt. The rotational speed ω may be detected using an appropriate speed sensor. Alternatively, the rotational speed Ne of the engine 3 detected by a rotational speed sensor (not shown) of the engine 3 may be used instead of the rotational angular speed ω.

  Here, the phase corrector 101 corresponds to the correction amount determining means in the present invention. Although details of the processing will be described later, the phase corrector 101 calculates and stores the magnetic pole correction angle θofs under a predetermined condition, and stores and holds it during normal operation of the motor 1 other than the predetermined condition. The magnetic pole correction angle θofs is output. The phase corrector 101 outputs a temporary setting value α of the magnetic pole correction angle θofs (hereinafter referred to as a temporary setting correction angle α) when executing the process of obtaining the magnetic pole correction angle θofs. In this case, the magnetic pole rotation angle θ obtained by the subtraction processor 19 is θ = θact−α.

  Torque is provided outside the motor control device 100 based on data representing the operation mode of the engine 3 (for example, flag data indicating whether or not the engine 3 is in the start mode) according to the operation of a start switch (not shown) of the vehicle. A vehicle control unit 50 that outputs a command Trc is provided. The torque command Trc output from the vehicle control unit 50 is input to the current command generator 9. Further, the vehicle control unit 50 controls the torque command Trc to be substantially zero and the rotational speed ω of the motor 1 to be a predetermined value or less and to be substantially constant, and indicates a magnetic pole position correction indicating whether or not to execute a magnetic pole position correction described later. Output permission signal. The magnetic pole position correction permission signal output from the vehicle control unit 50 is input to the phase corrector 101. Further, the vehicle control unit 50 outputs a signal instructing the number of revolutions of the engine 3 to the control unit 3 b that controls the operation of the engine 3.

  The phase corrector 101 receives the calculated values of the d-axis voltage command value Vdc and the q-axis voltage command value Vqc from the voltage command generator 9 and the calculated value of the rotational angular velocity ω from the speed calculator 20. . Further, the magnetic pole position correction permission signal output from the vehicle control unit 50 is input to the phase corrector 101. When the magnetic pole position correction permission signal is input to the phase corrector 101 and the process of obtaining the magnetic pole correction angle θofs is started, the d-axis current command value Idc and the q-axis current command value Iqc having the value “0” are output from the current command switch 10. A switching command for outputting the set (0, 0) is output to the current command switching unit 10.

  The phase corrector 101 divides the period in which the d-axis current command value Idc = 0 and the q-axis current command value Iqc = 0 are output from the current command switch 10 into two, and at a constant rotational speed that is different in each period. The vehicle control unit 50 is instructed to drive the engine 3. The phase corrector 101 records, in the memory 103, the rotation speed of the engine 3 and the magnetic pole correction angle θofs obtained by calculation in each period. In this embodiment, since the output shaft 3 a of the engine 3 and the rotor 4 of the motor 1 are connected, the rotational angular speed of the engine 3 is equal to the rotational angular speed of the rotor 4 of the motor 1. Therefore, the rotational angular velocity ω of the rotor 4 of the motor 1 may be recorded in the memory 103 instead of the rotational speed of the engine 3.

  Further, the phase corrector 101 determines the relationship between the rotational speed of the engine 3 and the magnetic pole correction angle θofs (hereinafter referred to as rotational speed−error) based on the rotational speed of the engine 3 and the magnetic pole correction angle θofs recorded in the memory 103. (Referred to as a characteristic). When the period during which the d-axis current command value Idc = 0 and the q-axis current command value Iqc = 0 is output is divided into two, the set of the engine speed and the magnetic pole correction angle θofs stored in the memory 103 Since there are two sets, the rotational speed-error characteristic is approximated to at least a straight line.

  FIG. 3 is a graph that approximates the number of times of calculation with respect to the magnetic pole correction angle θofs obtained in a state where the engine 3 is driven at a constant rotational speed of 1500 rpm. As shown in FIG. 3, the magnetic pole correction angle θofs when the engine 3 is driven at a constant rotation speed of 1500 rpm is −1.50 ° to −0.50 ° centered at −1.00 °. In most cases, a value between is required.

  FIG. 4 is a graph that approximates the number of times of calculation for the magnetic pole correction angle θofs obtained in a state where the engine 3 is driven at a constant rotational speed of 800 rpm. As shown in FIG. 4, the magnetic pole correction angle θofs when the engine 3 is driven at a constant rotation speed of 800 rpm is between + 0.50 ° and + 1.50 ° centered on + 1.00 °. In most cases, a value is required.

  FIG. 5 is a graph showing the distribution of the magnetic pole correction angle θofs calculated in a state where the engine 3 is driven at each rotational speed shown in FIGS. 3 and 4 and its approximate straight line. As shown in FIG. 5, a rotational speed-error characteristic approximated by a straight line is obtained from the relationship between two different rotational speeds and the magnetic pole correction angle θofs corresponding to each rotational speed.

The equation representing the approximate straight line is expressed as “y = Ax + B” where the magnetic pole correction angle θofs is a variable y and the rotational speed Nm of the engine 3 is a variable x. The coefficient A and the coefficient B are expressed by the following formulas.
Coefficient A = (Magnetic pole correction angle θofs1−Magnetic pole correction angle θofs2) ÷ (Rotational speed Nm1−Rotational speed Nm2)
Coefficient B = Magnetic pole correction angle θofs1-A × rotational speed Nm1

  If three or more different rotation speeds and the magnetic pole correction angle θofs corresponding to each rotation speed are obtained, the rotation speed-error characteristic can be approximated by a curve. Further, when the rotation speed-error characteristic is derived with a high load for the phase corrector 101, for example, a 10-point lookup table may be created and updated at this time until 10-point interpolation is performed.

  In the apparatus of the present embodiment described above, the current command generator 9, the current command switching device 10, the subtraction processors 13, 14, 19, the voltage command generator 15, the current coordinate converter 12, the voltage coordinate converter 16, the phase In this embodiment, the corrector 101 and the speed calculator 20 are configured as functional means of a microcomputer (including an input / output circuit) in which a predetermined program is installed.

  Next, the operation of the apparatus of this embodiment will be described with reference to the timing charts of FIGS. 6 and 7 and the flowcharts of FIGS. When a start switch (not shown) of the vehicle is turned on, power is supplied to the motor control device 100 and the like. Then, in order to start the engine 3 by operating the motor 1 as a starting motor of the engine 3, the motor control device 100 receives a torque command Trc from an arithmetic processing unit (not shown) as shown in the second stage diagram of FIG. (Time t1 to t2 in FIG. 6). At this time, the phase corrector 101 of the motor control device 100 outputs the magnetic pole correction angle θofs currently stored and held, and the current command switcher 10 receives the d-axis current command value Idc and the current command generator 9. A switching command for outputting the q-axis current command value Iqc is given. In this state, the motor control device 100 executes the above-described processes of the current command generator 9, the subtraction processors 13, 14, 19, the voltage command generator 15, the current coordinate converter 12, and the voltage coordinate converter 16. Thereby, the motor 1 is caused to generate torque (torque for starting the engine 3) corresponding to the torque command Trc. Thereby, cranking of the engine 3 is started. In addition, fuel supply control and ignition control of the engine 3 are performed by an engine controller (not shown), and the engine 3 is started.

  As the engine 3 starts, the rotational speed Ne of the engine 3 (in this embodiment, this is equal to the rotational angular speed ω of the motor 1) increases as shown in the fifth stage diagram of FIG. If it becomes above, the starting mode of the engine 3 will be complete | finished (idling driving | operation of the engine 3 will start), and the torque command Trc of the motor 1 will be "0" (time t2 of FIG. 6). Here, it is assumed that the accelerator operation of the vehicle is not performed.

  At this time, a start completion flag indicating whether or not the start of the engine 3 has been completed is set to "1" by an arithmetic processing unit (not shown) as shown in the fourth stage diagram of FIG. The start completion flag is input to the phase corrector 101 of the motor control device 100 as data representing the operation mode of the engine 3.

  Next, when the idling operation of the engine 3 is continued and the rotational speed Ne of the engine 3 or the rotational angular speed ω of the motor 1 is maintained substantially constant around a predetermined idling rotational speed (for example, 800 to 1000 rpm), the motor 1 Is set to "1" by the phase corrector 101 (time t3 in FIG. 6). The stability check flag indicating whether or not the rotation speed is equal to or less than a predetermined value (for example, 2000 rpm). At this time, since the magnetic pole position correction permission signal is input to the phase corrector 101, the phase corrector 101 executes a process for newly obtaining the magnetic pole correction angle θofs (time t3 to t4 in FIG. 6). ).

  Hereinafter, processing for newly obtaining the magnetic pole correction angle θofs will be described in detail with reference to FIG. FIG. 6 is a graph showing changes in the rotational speed, d-axis current, q-axis current, and voltage phase of the engine 3 when the motor control device 100 shown in FIG. 1 obtains the rotational speed-error characteristic. When the phase corrector 101 starts processing for newly obtaining the rotation speed-error characteristic, the value of “0” is output from the current command switch 10 while the engine 3 is rotating at a constant rotation speed of, for example, 1500 rpm. A switching command for outputting a set (0, 0) of the d-axis current command value Idc and the q-axis current command value Iqc is output to the current command switch 10. When the voltage phase is converged by outputting the switching command, the phase corrector 101 obtains the magnetic pole correction angle θofs. The phase corrector 101 records in the memory 103 the set of the rotation speed at this time and the obtained magnetic pole correction angle θofs.

  Next, the phase corrector 101 drives the engine 3 at a constant rotational speed (for example, 800 rpm) different from the rotational speed (1500 rpm) at the predetermined period D1 when the predetermined period D1 has elapsed from the switching command. The vehicle control unit 50 is instructed. The phase corrector 101 obtains the magnetic pole correction angle θofs when the voltage phase converges while the engine 3 is rotating at a constant rotational speed (800 rpm) indicated by the engine 3. The phase corrector 101 records in the memory 103 the set of the rotation speed at this time and the obtained magnetic pole correction angle θofs. The phase corrector 101 instructs the vehicle control unit 50 to stop driving the engine 3 when a predetermined period D2 has elapsed from the rotation speed switching instruction.

  The processing of the phase corrector 101 in the sequence as described above is executed as shown in the flowcharts of FIGS.

  In step S101, the phase corrector 101 determines whether or not it is immediately after the start of the engine 3 is completed based on the value of the start completion flag (see FIG. 6). If the determination result is YES (time t2 in FIG. 6), the phase corrector 101 further determines whether the rotational speed ω of the motor 1 is equal to or lower than a first predetermined value (for example, 1250 rpm). Are determined in steps S103 and S105, respectively. The determination in step S105 is, for example, whether or not the temporal fluctuation width of the rotational speed ω of the motor 1 obtained by the speed calculator 20 is within a predetermined time and not more than a predetermined width δ (see the fifth stage diagram in FIG. 6). It is done by judging. If both of these determination results are YES, the phase corrector 101 further determines in step S107 whether or not the torque command value Trc is substantially “0”. This determination is made by determining whether or not the torque command value Trc is within a predetermined range near “0”.

  When the determination result in step S107 is YES (when all the determination results in steps S101 to S107 are YES), the phase corrector 101 executes the process in step S109 to set the magnetic pole correction angle θofs to the following. Ask to explain. Moreover, a process is complete | finished when the judgment result in any one of step S101-S107 is NO. In this case, the phase corrector 101 outputs the magnetic pole correction angle θofs corresponding to the rotational speed based on the currently held rotational speed-error characteristic.

  In step S109, the phase corrector 101 sends a switching command for causing the current command switching unit 10 to output the d-axis current command value Id = 0 and the q-axis current command value Iq = 0 having the value “0” to the current command switching unit 10. give. Next, in step S <b> 111, the phase corrector 101 calculates the magnetic pole correction angle θofs <b> 1, and records the rotation speed Nm <b> 1 of the engine 3 at this time and the calculated magnetic pole correction angle θofs <b> 1 in the memory 103. Details of the processing performed in step S111 are shown in FIG. As shown in FIG. 10, in step S201, the phase corrector 101 sets the temporarily set correction angle α to “0” and outputs this instead of the magnetic pole correction angle θofs.

  In this case, the magnetic pole rotation angle θ obtained by the subtraction processor 19 is θ = θact−α = θact. Therefore, the magnetic pole detection angle θact by the magnetic pole position detector 8 is directly input to the converters 12 and 16 as the magnetic pole rotation angle θ used in the coordinate conversion of the current coordinate converter 12 and the voltage coordinate converter 16. Become. In this state, the motor control device 100 obtains the phase voltage command values Vuc, Vvc, Vwc so that the d-axis detection current Id and the q-axis detection current Iq match the command values “0”. The voltage applied to the armature of the motor 1 is manipulated. As a result, the actual armature current (current flowing through each phase of U, V, W) of the motor 1 is controlled to be substantially “0”.

  After executing the process of step S201, the phase corrector 101 waits for a predetermined time in step S203, and then executes the process of step S205 described later. Here, the predetermined time is a time determined in advance as a necessary and sufficient time until the actual armature current of the motor 1 sufficiently converges to the vicinity of “0” after the process of step S109 is executed. For example, 0.5 seconds. Instead of waiting for a predetermined time in step S203, the armature currents detected by the armature current detectors 11u and 11v are monitored, and when they are within a predetermined range near “0”, step S205 is performed. You may make it perform the process of.

  In step S203, the state of the motor 1 becomes a zero current state. At this time, the rotation angle of the magnetic pole recognized in the dq vector control processing of the motor control device 100 (the rotation angle position of the command axis dc in FIG. 14B), that is, the magnetic pole rotation angle θ is Since it is the magnetic pole detection angle θact by the position detector 8, the error angle of the magnetic pole detection angle θact with respect to the actual rotation angle of the magnetic pole position is expressed by Expression (1). Therefore, in step S205, the phase corrector 101 calculates the right side of the equation (1) using the d-axis voltage command value Vdc and the q-axis voltage command value Vqc obtained by the voltage command generator 15, and the calculation is performed. The value obtained by the above is newly obtained as the magnetic pole correction angle θofs1. Further, the phase corrector 101 records the obtained magnetic pole correction angle θofs1 and the rotation speed Nm1 of the engine 3 in the memory 103 in step S207.

  Next, the phase corrector 101 instructs the vehicle control unit 50 to switch the constant rotational speed of the engine 3 in step S113. Next, the phase corrector 101 performs steps S115 to S119, which are the same processes as steps S103 to S107. However, in step S115, the phase corrector 101 determines whether or not the rotational speed ω of the motor 1 is a second predetermined value (for example, 850 rpm). After performing Step S119, the phase corrector 101 performs Step S121, which is the same processing as Step S111. In step S 121, the phase corrector 101 calculates the magnetic pole correction angle θofs 2, and records the rotation speed Nm 2 of the engine 3 at this time and the calculated magnetic pole correction angle θofs 2 in the memory 103. Details of the processing performed in step S121 are the same as the processing shown in FIG.

  Next, the phase corrector 101, based on the rotation speed Nm1 and the magnetic pole correction angle θofs1 recorded in the memory 103 in step S111, and the rotation speed Nm2 and the magnetic pole correction angle θofs2 recorded in the memory 103 in step S121, The relationship between the rotational speed of the engine 3 and the magnetic pole correction angle θofs (hereinafter referred to as rotational speed-error characteristic) is obtained. The rotation speed-error characteristic obtained at this time is recorded in the memory 103.

  Thereafter, in step S125, the phase corrector 101 performs the process of obtaining the rotation speed-error characteristic again (the process of steps S101 to S123), that is, until the next operation of the engine 3 is resumed. (However, during driving of the vehicle), the magnetic pole correction angle θofs corresponding to the rotational speed of the engine 3 is output based on the rotational speed-error characteristic recorded in the memory 103 in step S123. Therefore, hereinafter, the magnetic pole rotation angle θ obtained by subtracting the magnetic pole correction angle θofs from the magnetic pole detection angle θact represents the actual magnetic pole position of the rotor 4 and is used for controlling the motor 1 by the motor control device 100. It is done.

  As described above, the magnetic pole correction angle θofs obtained in step S125 represents the error angle of the magnetic pole detection angle θact with respect to the actual rotation angle of the magnetic pole position in accordance with the principle described with reference to FIG. In this case, in this embodiment, it is necessary to perform field weakening control when the rotational speed ω of the motor 1 is equal to or lower than a predetermined value, that is, the d-axis current command value Idc is intentionally set to a negative value. A process for obtaining the magnetic pole correction angle θofs is executed in the absence of the magnetic field. Further, the magnetic pole correction angle θofs is obtained in a state where the rotational speed ω of the motor 1 is substantially constant, that is, in a state where the back electromotive force generated in each phase of the armature 7 is substantially constant. For this reason, the magnetic pole correction angle θofs can be obtained with high accuracy. This is the same not only when the permanent magnet 5 of the rotor 4 is cylindrical, but also when it is salient.

  As described above, according to the present embodiment, the magnetic pole correction angle θofs is calculated in each state where the engine 3 is rotating at a plurality of different constant rotational speeds, and the rotational speed-error characteristics are obtained. The magnetic pole correction angle θofs corresponding to the number of rotations of 3 is output. Accordingly, it is possible to derive the magnetic pole correction angle θofs that can correct the error depending on the number of rotations included in the magnetic pole detection position, the mounting error of the magnetic pole position detector 8, the error due to aging, and the error due to product variations.

  In this embodiment, in step S205 of FIG. 10, the magnetic pole correction angle θofs is obtained by equation (1), but may be obtained by equation (2) or (3). In the present embodiment, the temporarily set correction angle α used during the process of obtaining the magnetic pole correction angle θofs is set to “0”, but other values, for example, the current value of the magnetic pole correction angle θofs are temporarily set. It may be used as the setting correction angle α. In this case, in step S205 of FIG. 10, the value obtained by the calculation of the right side of the equation (1), the equation (2), or the equation (3) is added to the current value of the magnetic pole correction angle θofs to obtain a new value. A simple magnetic pole correction angle θofs may be obtained.

  The details of the processing performed in step S111 and step S121 in the flowcharts shown in FIGS. 8 and 9 are shown in FIG. 10, but the processing shown in FIG. 11 may be performed in this step. That is, as shown in FIG. 11, the phase corrector 101 sets the initial value of the temporarily set correction angle α to “0”, for example, in step S211, and outputs this instead of the magnetic pole correction angle θofs, and then in step S213. Wait for a predetermined time. As in the case of step S203 in FIG. 10, the predetermined time is a time (for example, 0.5 times) determined in advance as a necessary and sufficient time until the actual armature current of the motor 1 sufficiently converges to the vicinity of “0”. Second). Instead of waiting for a predetermined time in step S213, the armature currents detected by the armature current detectors 11u and 11v are monitored, and when they fall within a predetermined range near “0”, You may make it perform the process of step S215.

  Next, in step S215, the phase corrector 101 determines whether or not the d-axis voltage command value Vdc obtained by the voltage command generator 15 is substantially “0”. This determination is made by determining whether or not Vdc is within a predetermined range near “0”. At this time, if the determination result in step S215 is YES, the magnetic pole rotation angle θ obtained by subtracting the temporarily set correction angle α currently output by the phase corrector 101 from the magnetic pole detection angle θact is the rotor 4 The actual rotation angle of the magnetic pole is substantially the same (the state shown in FIG. 14A).

  Therefore, in this case, in step S217, the phase corrector 101 obtains the currently set temporarily set correction angle α as the magnetic pole correction angle θofs. In step S 219, the magnetic pole correction angle θofs and the engine speed Nm are recorded in the memory 103.

  On the other hand, if the determination result in step S215 is NO, the magnetic pole rotation angle θ obtained by subtracting the temporarily set correction angle α currently output by the phase corrector 101 from the magnetic pole detection angle θact is It does not coincide with the actual rotation angle of the magnetic pole (the state shown in FIG. 14B). In this case, in step S221, the phase corrector 101 updates the temporary setting correction angle α to a value obtained by increasing the current value by a small predetermined amount Δα, and then from step S213. Repeat the process.

  In this way, the value of the provisional correction angle α is changed by a predetermined amount Δα until the d-axis voltage command value Vdc becomes substantially “0”, and the d-axis voltage command value Vdc becomes almost “0”. The temporarily set correction angle α is searched. Then, the searched value of the temporarily set correction angle α is finally obtained as the magnetic pole correction angle θofs and recorded in the memory 103.

  In the example shown in FIG. 11, the initial value of the temporary setting correction angle α is set to “0”. However, for example, the current value of the magnetic pole correction angle θofs is set as the initial value of the temporary setting correction angle α, and the initial value The temporarily set correction angle α may be changed in the vicinity range.

  In the embodiment described above, the magnetic pole correction angle θofs is obtained when the torque command value Trc (requested torque) of the motor 1 is substantially “0” during the idling operation immediately after the engine 3 is started. However, for example, the magnetic pole correction angle θofs may be obtained during idling while the vehicle is temporarily stopped. Further, in a situation where the motor 1 does not need to generate torque, for example, the magnetic pole correction angle θofs when the vehicle is cruising. May be requested. In this embodiment, for example, the determination process in step S101 in FIG. 8 may be omitted.

  In the present invention, basically, the magnetic pole correction angle θofs can be obtained in a situation where the motor 1 does not need to generate torque. For example, the motor 1 runs idle without generating torque. When driving, that is, when the vehicle is coasting without setting the accelerator operation amount to “0” during traveling and applying driving force to the driving wheel, the magnetic pole correction angle θofs is obtained. Also good. For example, the magnetic pole correction angle θofs is obtained if the rotational speed of the rotor 4 is substantially constant even when the rotational speed is decreasing due to rotational friction of the motor 1. In this embodiment, for example, instead of the determination process in step S101 of FIG. 8, it is determined whether or not the accelerator operation of the vehicle is OFF (whether or not the accelerator operation amount is “0”). When the determination result is YES, the process proceeds to the next step.

  In any case, it is preferable to obtain the magnetic pole correction angle θofs while the rotational speeds of the engine 3 and the motor 1 are stable. That is, when the rotational speed of the rotor 4 is substantially constant, the back electromotive force E of the armature is substantially constant, so that the d-axis voltage command value Vdc and the q-axis voltage command value Vqc in the zero current state are also substantially constant. Become. Therefore, the relational expressions (1) to (3) are established with high accuracy. Therefore, a highly reliable (accurate) magnetic pole correction angle θofs can be obtained.

  Moreover, it is preferable to perform the process of obtaining the magnetic pole correction angle θofs when the rotational speed of the rotor 4 is a predetermined speed (for example, 2000 rpm) or less. That is, in the high-speed rotation region of the motor 1, the back electromotive force E of the armature 4 becomes large. Therefore, in order to make the armature current I substantially zero, the d-axis current is intentionally set to a negative value, It is necessary to control the motor 1 that weakens the field (so-called field weakening control). In this state, the relational expressions (1) to (3) do not hold. Therefore, by performing the process of obtaining the magnetic pole correction angle θofs when the rotational speed of the rotor 4 is equal to or lower than the predetermined speed, the reliability of the magnetic pole correction angle θofs obtained by the process can be ensured.

  In the present embodiment, the magnetic pole correction angle θofs is basically obtained every time the engine 3 is started. For example, when the production of the vehicle at the vehicle production factory is completed, or maintenance of the vehicle is performed. The magnetic pole correction angle θofs may be obtained and updated at the time of inspection. In this embodiment, for example, an operator at a production factory or a maintenance / inspection worker previously sets an operation switch for instructing whether or not to permit the phase corrector 101 of the motor control device 100 to calculate the magnetic pole correction angle θofs. A process for determining ON / OFF of the operation switch is executed before the determination process of step S101 in FIG. Then, the phase corrector 101 only needs to execute the processing from step S101 in FIG. 8 only when the operation switch is turned ON by the operator.

  In the present embodiment, the control of the motor 1 mounted on the parallel hybrid vehicle has been described. However, the present invention is also applied to, for example, a permanent magnet type rotating electrical machine mounted on a series hybrid vehicle as a driving motor. Of course you can. Furthermore, the present invention can also be applied to permanent magnet type rotating electrical machines (permanent magnet type electric motors or permanent magnet type electric generators) used as prime movers other than vehicles.

  Further, in this embodiment, the current command switching device 10 is provided, and it is determined in step S107 whether or not the torque command is substantially “0”, and then the process proceeds to step S109. When the torque command can be clearly confirmed to be “0” as in the case where the present invention is applied to a permanent magnet generator mounted on a power generator (a generator having an engine as a drive source), the current command switch 10 Can be omitted.

  In the present embodiment, the motor 1 is controlled by the dq vector control. The actual operation control of the motor 1 itself is performed by a control method other than the dq vector control, and the magnetic pole correction angle θofs. It is also possible to ask for. In this case, for example, when obtaining the magnetic pole correction angle θofs, the phase voltage command values Vu, Vv, Vw are input to the PWM inverter circuit 17 from another controller that performs control other than the dq vector control, and the motor 1 While current control is performed so that the armature current is set to “0”, the d-axis voltage command value Vdc and the q-axis voltage command value Vqc are obtained by the dq vector control process similar to that of the motor control device 100 of the present embodiment. (However, these command values Vdc and Vqc are not used to actually control the motor 1). As in the present embodiment, the magnetic pole correction angle θofs may be obtained.

The block diagram which shows the apparatus structure of embodiment of this invention. 1 is a cross-sectional view schematically showing a permanent magnet type rotating electrical machine of the apparatus of FIG. A graph approximating the number of calculations for the magnetic pole correction angle θofs obtained when the engine 3 is driven at a constant rotational speed of 1500 rpm. A graph approximating the number of calculations for the magnetic pole correction angle θofs obtained when the engine 3 is driven at a constant rotational speed of 800 rpm. A graph showing the distribution of the magnetic pole correction angle θofs calculated in a state where the engine 3 is driven at each rotational speed shown in FIGS. 3 and 4 and its approximate straight line Timing chart for explaining the operation of the embodiment of the present invention 1 is a graph showing changes in the rotational speed, d-axis current, q-axis current, and voltage phase of the engine 3 when the motor control device 100 shown in FIG. 1 calculates the magnetic pole correction angle θofs. The flowchart for demonstrating the action | operation of the phase corrector in one Embodiment of this invention. The flowchart for demonstrating the action | operation of the phase corrector in one Embodiment of this invention. The flowchart which shows the detail of the process performed by step S111, S121 The flowchart which shows the detail of the process performed by step S111, S121 FIG. 2 is a block diagram showing an internal configuration of a control device for a permanent magnet type rotating electrical machine disclosed in Patent Document 2 in which a rotor is connected to an output shaft of an engine. 12 is a graph showing changes in engine speed, d-axis current, q-axis current, and voltage phase when the control device shown in FIG. 12 calculates the magnetic pole position correction amount. The figure for demonstrating the principle of this invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Motor 100 Motor control apparatus 3 Engine 4 Rotor 5 Permanent magnet 6 Stator 7 Armature 8 Magnetic pole position detector 9 Current command generator 10 Current command switch 11u, 11v Current detector 12 Current coordinate converter 13, 14 Subtraction Processor 15 Voltage command generator 16 Voltage coordinate converter 17 PWM inverter circuit 101 Phase corrector 103 Memory 50 Vehicle controller

Claims (3)

A magnetic pole position detection unit for detecting a magnetic pole position of the rotor of a permanent magnet type rotating electric machine in which a permanent magnet and an armature are provided on the rotor and the stator, respectively;
An error calculation unit that calculates an error of the magnetic pole position detected by the magnetic pole position detection unit with respect to the actual magnetic pole position in a state where the rotor is rotating at a substantially constant rotational speed;
A plurality of errors calculated by the error calculation unit each time the rotor is rotating at a substantially constant rotation number, and the rotation speed of the rotor when the error calculation unit calculates each error. A characteristic deriving unit for deriving a rotational speed-error characteristic indicating a relationship between the rotational speed of the rotor and an error in magnetic pole position,
A correction amount determination unit that obtains a magnetic pole position correction amount according to the rotation number of the rotor for correcting the magnetic pole position detected by the magnetic pole position detection unit based on the rotation speed-error characteristic;
While manipulating the phase of the armature voltage applied to the armature according to the magnetic pole position obtained by correcting the magnetic pole position detected by the magnetic pole position detection unit with the magnetic pole position correction amount obtained by the correction amount determination unit In the control device for the permanent magnet type rotating electrical machine, comprising: a current control unit that controls an armature current flowing through the armature;
In each state where the rotor is rotating at two or more different substantially constant rotational speeds, the error calculation unit has a field direction of the rotor as a d-axis and a direction perpendicular to the d-axis as a q-axis. While maintaining both the d-axis current command value and the q-axis current command value in the dq vector control for handling the rotating electrical machine in the dq coordinate system to be zero, the magnetic pole position correction amount is set to a predetermined temporary setting value and the dq A predetermined calculation using the d-axis voltage command value and the q-axis voltage command value as variables from the d-axis voltage command value and the q-axis voltage command value obtained when the vector control process is executed. A control device for a permanent magnet type rotating electrical machine, wherein an error is calculated based on an equation. A control device for a permanent magnet type rotating electrical machine according to claim 1,
When the error calculation unit calculates the magnetic pole position error in each state where the rotor rotates at two different substantially constant rotation numbers, the characteristic deriving unit calculates the rotation speed-error characteristic by linear approximation. A control device for a permanent magnet type rotating electrical machine, wherein the controller is derived. A control device for a permanent magnet type rotating electrical machine according to claim 1,
When the error calculation unit calculates the magnetic pole position error in each state where the rotor rotates at three or more different substantially constant rotation numbers, the characteristic deriving unit calculates the rotation speed-error by curve approximation. A control device for a permanent magnet type rotating electrical machine characterized by deriving characteristics.

Images