JP2010029030A - Motor controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a motor controller capable of executing estimation calculation of a rotation angle, only as required, obtaining a proper estimated rotation angle rapidly, after the start of calculation and thus properly controlling a motor, while reducing the calculation load. <P>SOLUTION: A sensor failure judging section 25 determines that there is no failure of a resolver 2. At normal times when there is no failure in the resolver 2, the motor 1 is controlled using a detected rotation angle θ<SB>S</SB>detected by the resolver 2. During this time, a rotation angle estimating section 31 stops its calculation for estimation. If there is a failure in the resolver 2, a search voltage command value is generated from a search voltage generating section 26 and a stator of the motor 1 forms a search magnetic field. An initial value determining section 24 determines an initial value of an internal variable of the rotation angle estimating section 31, based on the output of a toque sensor 7 then. By using this initial value, the rotation angle estimating section 31, an estimated rotation angle θ<SB>E</SB>, is obtained. The motor 1 is controlled by using this estimated rotation angle θ<SB>E</SB>. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a motor control device for driving a brushless motor. The brushless motor is used, for example, as a source for generating a steering assist force in an electric power steering apparatus.

A motor control device for driving and controlling a brushless motor is generally configured to control the supply of motor current in accordance with the output of a rotation angle sensor for detecting the rotation angle of the rotor. As the rotation angle sensor, for example, a resolver that outputs a sine wave signal and a cosine wave signal corresponding to the rotor rotation angle (electrical angle) is used.
If a rotation angle sensor failure (including a signal line disconnection failure or a short-circuit failure) occurs, it becomes impossible to specify the rotor rotation angle, so that the drive control of the brushless motor cannot be continued.

This problem is alleviated by the combined use of a sensorless drive system that drives a brushless motor without using a rotation angle sensor. In the sensorless drive system, for example, a rotation angle estimator that estimates the phase of a magnetic pole (electrical angle of the rotor) by estimating an induced voltage associated with the rotation of the rotor (Patent Document 1). The rotation angle estimator is, for example, function processing means that is realized by a microcomputer executing a predetermined program.
JP 2004-7924 A Masashi Ichikawa et al., “Sensorless control of salient-pole permanent magnet synchronous motor based on the extended induced voltage model”, IEEJ Transactions D, 122, 12, 2002

  The rotation angle estimator usually has an internal variable including information on past motor current and motor voltage (or voltage command value). The rotation angle estimator is configured to estimate the rotor rotation angle and update internal variables. Therefore, by repeating the rotor rotation angle estimation calculation at a predetermined calculation cycle, the internal variable converges from its initial value to a value according to the current state of the motor, and accordingly the estimated rotation angle also changes to a reasonable value. And converge.

  The initial value of the internal variable is normally a constant value (for example, zero), and therefore the estimated rotation angle in the initial period when the rotation angle estimator starts to operate is unstable. Therefore, when the rotation angle sensor and the sensorless driving method are used in combination, even when the rotation angle sensor is normal and therefore the estimated rotation angle is not required, the calculation of the rotation angle estimator is performed in parallel. It is necessary to converge the estimated rotation angle in advance to an appropriate value.

However, if the rotation angle estimator is operated until the estimation of the rotation angle is unnecessary, the calculation load increases. For this reason, a computer used for the motor control device needs to have a high computing capacity, and the cost increases accordingly.
Therefore, an object of the present invention is to perform the rotation angle estimation calculation only when necessary, and to obtain a reasonable estimated rotation angle immediately after the calculation starts, thereby reducing the calculation load. A motor control device capable of appropriately controlling a motor is provided.

The invention described in claim 1 for achieving the above object is to provide a motor control device (10) for controlling a motor (1) including a rotor (50) and a stator (55) facing the rotor. A rotation angle detection means (2, 22) for detecting the rotation angle of the rotor, a rotation angle estimation means (31) for estimating the rotation angle of the rotor, and a torque generated by the motor A torque detection means (7) for detecting the rotation angle, a search magnetic field generation means (26) for generating a search magnetic field from the stator when the rotation angle detection means fails, and the torque detection when the search magnetic field is generated An initial value determining means (24) for determining an initial value for an estimation calculation in the rotation angle estimating means based on a detected torque detected by the means; and a normal time when the rotation angle detecting means is not faulty. Controlling the motor based on the detected rotation angle rotation angle detecting means for detecting (θ _S), the estimated rotational angle at the time of failure the rotation angle estimating means has estimated on the basis of the initial value the rotation angle detecting means (theta _E ), and a control means (15-21, 32) for controlling the motor based on the motor control device. The alphanumeric characters in parentheses indicate corresponding components in the embodiments described later. The same applies hereinafter.

  According to this configuration, an exploration magnetic field is generated from the stator when the rotation angle detection means fails. When torque is generated in the rotor by the exploration magnetic field, this torque is detected by the torque detection means. Based on this detected torque, an initial value for the estimation calculation in the rotation angle estimation means is determined. Since the torque response when the exploration magnetic field is applied varies depending on the rotation angle of the rotor, the initial value is determined by reflecting the rotation angle of the rotor. Using the initial value determined in this way, an estimation calculation is performed by the rotation angle estimation means, and an estimated rotation angle is obtained. Since the initial value is a value reflecting the rotor rotation angle, the estimated rotation angle quickly converges to an appropriate value. Therefore, in consideration of the waiting time for convergence of the estimated rotation angle, it is not necessary to start the estimation calculation by the rotation angle estimating means in advance when the estimated rotation angle is unnecessary.

  Therefore, when the motor is controlled based on the detected rotation angle detected by the rotation angle detection means without any failure in the rotation angle detection means, the estimation calculation by the rotation angle estimation means should be stopped. Can do. Then, when a failure occurs in the rotation angle detection means, the estimation calculation by the rotation angle estimation means is started, and the initial value obtained as described above may be applied to the estimation calculation. Thereby, even if the rotation angle estimation calculation is started after it is determined that the rotation angle detecting means has failed, a reasonable estimated rotation angle can be obtained quickly, and the motor can be driven appropriately. .

Thus, since the motor can be appropriately controlled while reducing the calculation load, the cost of the motor control device can be reduced.
Preferably, the motor control device further includes a failure determination unit (25) for determining whether or not a failure has occurred in the rotation angle detection unit. In this case, the exploration magnetic field generating means preferably generates the exploration magnetic field from the stator in response to the failure determination means determining that the rotation angle detection means has failed. The control means preferably controls the motor by selecting either the detected rotation angle or the estimated rotation angle based on the determination result by the failure determination means.

  The control means may include voltage command value generation means (19A, 19B) for generating a motor voltage command value. In this case, the motor control device may further include drive means (13) for driving the motor according to the motor voltage command value and current detection means (11) for detecting the motor current. In this case, the rotation angle estimation means may estimate the rotation angle of the rotor based on the motor voltage command value and the motor current.

  The exploration magnetic field generating means preferably applies the exploration magnetic field one after another while changing the direction. Thereby, the said initial value can be determined based on the response of the motor torque when the exploration magnetic field is applied in a plurality of directions. Therefore, since the initial value can be determined more accurately, the estimated rotation angle can be quickly converged to a reasonable value accordingly. In this case, it is preferable that the exploration magnetic field generating means sequentially changes the direction of the exploration magnetic field in an order that suppresses rotation of the rotor due to application of the exploration magnetic field. More specifically, the exploration magnetic field generating means preferably generates the exploration magnetic field from the stator while sequentially changing the direction so that the phase is substantially reversed (differing by 180 °) according to the electrical angle of the rotor. Thus, the initial value can be determined while suppressing the rotation of the rotor due to the exploration magnetic field.

According to a second aspect of the present invention, the exploration magnetic field generating means sequentially generates a plurality of exploration magnetic fields in different directions, and the initial value determining means includes a zero-cross point of detection torque (detection torque with respect to the direction of exploration magnetic field). 2. The motor control device according to claim 1, wherein the initial value is determined based on a direction of the exploration magnetic field corresponding to a zero-crossing point).
By generating a plurality of search magnetic fields in different directions from the stator, the magnitude and direction of the torque generated by the motor when the search magnetic field is applied are changed. That is, the magnitude of the torque is minimized when the direction of the exploration magnetic field and the magnetic pole direction of the rotor are parallel, and the magnitude of the torque is maximized when the direction of the exploration magnetic field and the magnetic pole direction of the rotor are orthogonal. The direction of the torque depends on the magnetic pole position of the rotor.

  Accordingly, in the invention of claim 2, a zero cross point of the detected torque, that is, a point where the direction of the detected torque is reversed is found, and the initial torque is determined based on the direction of the search magnetic field at this zero cross point (a direction parallel to the rotor magnetic pole direction). The value is determined. More specifically, if a positive sign is given to the detected torque in one direction and a negative sign is given to the detected torque in the opposite direction, the point where the sign of the detected torque is inverted is the zero cross point. Whether the zero cross point is inverted from the positive sign to the negative sign or the zero cross point is inverted from the negative sign to the positive sign can be determined from the detected torque in the vicinity of the zero cross point.

According to a third aspect of the present invention, the exploration magnetic field generating means sequentially generates a plurality of exploration magnetic fields in different directions, and the initial value determining means is a maximum or minimum point of detected torque (the direction of the exploration magnetic field). The motor control device according to claim 1, wherein the initial value is determined based on a direction of an exploration magnetic field corresponding to a maximum or minimum point of a detected torque change with respect to.
At the maximum or minimum point of the detected torque, the direction of the search magnetic field and the rotor magnetic pole direction are orthogonal. Accordingly, the initial value can be determined from the direction of the search magnetic field at the maximum or minimum point of the detected torque. When a positive sign is given to the detected torque in one direction and a negative sign is given to the detected torque in the opposite direction, the detected torque has a positive sign at the maximum point, and the detected torque has a negative sign at the minimum point.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 is a block diagram for explaining an electrical configuration of an electric power steering apparatus to which a motor control apparatus according to an embodiment of the present invention is applied. This electric power steering apparatus includes a torque sensor 7 that detects an operation torque applied to a steering wheel of a vehicle, a vehicle speed sensor 8 that detects the speed of the vehicle, a motor 1 that applies a steering assist force to the steering mechanism 3 of the vehicle, And a motor control device 10 for driving and controlling the motor 1. The motor control device 10 drives the motor 1 in accordance with the operation torque detected by the torque sensor 7 and the vehicle speed detected by the vehicle speed sensor 8, thereby realizing appropriate steering assistance according to the steering situation. The motor 1 is, for example, a three-phase brushless motor, and, as schematically shown in FIG. 2A, a rotor 50 as a field and U-phase, V-phase, and W-phase stator windings 51, 52, And a stator 55 including 53. The motor 1 may be an inner rotor type in which a stator is arranged outside the rotor, or may be an outer rotor type in which a stator is arranged inside a cylindrical rotor.

The motor control device 10 includes a current detection unit 11, a microcomputer 12 as a signal processing unit, and a drive circuit 13. The above-described torque sensor 7 and vehicle speed sensor 8 are connected to the motor control device 10 together with the resolver 2 (rotation angle sensor) that detects the rotation angle of the rotor in the motor 1.
The current detector 11 detects the current flowing through the stator windings 51, 52, 53 of the motor 1. More specifically, the current detection unit 11 includes current detectors that respectively detect phase currents in the three-phase (U-phase, V-phase, and W-phase) stator windings 51, 52, and 53.

  The microcomputer 12 includes a CPU and a memory (such as a ROM and a RAM), and functions as a plurality of function processing units by executing a predetermined program. The plurality of function processing units include a basic target current value calculation unit 15, a dq-axis target current value calculation unit 16, a PI (proportional integration) control unit 19A, a voltage command value generation unit 19B, and γδ / αβ coordinates. Conversion unit 20A, αβ / UVW coordinate conversion unit 20B, PWM control unit 21, UVW / αβ coordinate conversion unit 17A, αβ / γδ coordinate conversion unit 17B, deviation calculation unit 18, and rotation angle calculation unit 22 The rotation angular velocity calculation unit 23, the initial value determination unit 24, the sensor failure determination unit 25, the exploration voltage generation unit 26, the switch 30, the rotation angle estimation unit 31, and the switching unit 32 are provided.

The basic target current value calculation unit 15 calculates the basic target current value I ^* of the motor 1 based on the operation torque detected by the torque sensor 7 and the vehicle speed detected by the vehicle speed sensor 8. For example, the basic target current value I ^* is determined so as to increase as the operating torque increases and to increase as the vehicle speed decreases.
Based on the basic target current value I ^* , the dq-axis target current value calculation unit 16 applies the d-axis current component target value (d-axis target current value I _d ^* ) along the rotor magnetic pole direction of the motor 1 to the d-axis. A target value (q-axis target current value I _q ^* ) of the orthogonal q-axis current component is generated. Below, to when referring collectively these are referred to as "target current value I _dq".

When the basic target current value I ^* representing the amplitude of the current (sine wave current) to be applied to the U phase, V phase and W phase of the motor 1 is used, the d axis target current value I _d ^* and the q axis target current value I _q ^* Is expressed by the following equations (1) and (2).

したがって、ｄｑ軸目標電流値演算部１６は、ｄ軸目標電流値Ｉ_d ^*＝０を生成する一方で、トルクセンサ７によって検出される操作トルクに応じたｑ軸目標電流値Ｉ_q ^*を生成する。

Therefore, dq-axis target current value calculation unit 16, while generating a d-axis target current value I _d ^* = 0, produces a q-axis target current value I _q ^* in accordance with the operation torque detected by the torque sensor 7 To do.

Current detecting unit 11, the U-phase current I _u motor 1 to detect the V-phase current I _v and the W-phase current I _w (hereinafter, when referred to collectively the "three-phase detected current I _uvw"). The detected value is given to the UVW / αβ coordinate converter 17A.
The UVW / αβ coordinate converter 17A converts the three-phase detection current I _uvw into the currents I _α and I _β on the two-phase fixed coordinate system (α-β) (hereinafter referred to as “two-phase detection current I”). _The coordinates are converted to “ _αβ ”. The two-phase fixed coordinate system (α-β) is a fixed coordinate system in which the rotation center of the rotor 50 is the origin and the α axis and the β axis orthogonal to the rotation axis of the rotor 50 are defined (see FIG. 2). ). The coordinate-converted two-phase detection current I _αβ is given to the αβ / γδ coordinate conversion unit 17B.

The αβ / γδ coordinate converter 17B converts the two-phase detection current I _αβ on the two-phase rotational coordinate system (γ-δ) according to the rotor rotation angle θ ^ on control (hereinafter referred to as “control rotation angle θ ^”). The coordinates are converted into currents I _γ and I _δ (hereinafter referred to as “two-phase detection current I _γδ ”). The two-phase rotational coordinate system (γ−δ) is a rotational coordinate system defined by a γ axis along the rotor magnetic pole direction and a δ axis orthogonal to the γ axis when the rotor 50 is at the control rotation angle θ ^. It is. When the control rotation angle θ ^ has no error and matches the actual rotor rotation angle, the two-phase rotation coordinate system (dq) and the two-phase rotation coordinate system (γ-δ) match. The control rotation angle θ ^ is a rotor rotation angle calculated by the rotation angle calculation unit 22 or the rotation angle estimation unit 31 and selected by the switching unit 32.

The two-phase detection current I _γδ is supplied to the deviation calculation unit 18. The deviation calculator 18 calculates a deviation of the γ-axis current I _{γ with} respect to the d-axis target current value I _d ^* and a deviation of the δ-axis current I _{δ with} respect to the q-axis target current value I _q ^* . These deviations are given to the PI control unit 19A and undergo PI processing. Then, depending on the results of these calculations, the voltage command value generating unit 19B, gamma-axis voltage command value V _gamma ^* and [delta] -axis voltage value V _[delta] ^* (hereinafter when referred to collectively as "two-phase voltage command value V _{the ??} "hereinafter.) is produced, provided to the ?? / .alpha..beta coordinate conversion unit 20A.

The γδ / αβ coordinate conversion unit 20A converts the γ-axis voltage command value V _γ ^* and the δ-axis voltage command value V _δ ^* into an α-axis voltage command value V that is a voltage command value of the two-phase fixed coordinate system (α-β). Coordinates are converted to _α ^* and β-axis voltage command value V _β ^* (hereinafter, collectively referred to as “two-phase voltage command value V _αβ ”). The two-phase voltage command value V _αβ is given to the αβ / UVW coordinate conversion unit 20B.

The αβ / UVW coordinate conversion unit 20B converts the α-axis voltage command value V _α ^* and the β-axis voltage command value V _β ^* into voltage command values of a three-phase fixed coordinate system, that is, voltage commands for the U phase, V phase, and W phase. The values are converted into values V _u ^* , V _v ^* , V _w ^* (hereinafter, collectively referred to as “three-phase voltage command value V _uvw ”).
The PWM control unit 21 generates a drive signal having a duty ratio controlled according to the three-phase voltage command values V _u ^* , V _v ^* , and V _w ^* and supplies the drive signal to the drive circuit 13. As a result, a voltage is applied to each phase of the motor 1 with a duty ratio corresponding to the voltage command values V _u ^* , V _v ^* , V _w ^* of the corresponding phase.

With such a configuration, when an operation torque is applied to a steering wheel (not shown) as an operation member coupled to the steering mechanism 3, this is detected by the torque sensor 7. Then, a target current value I _dq corresponding to the detected operation torque and vehicle speed is generated by the dq axis target current value calculation unit 16. A deviation between the target current value I _dq and the two-phase detection current I _γδ is obtained by the deviation calculation unit 18, and PI calculation is performed by the PI control unit 19 A so as to lead this deviation to zero. A two-phase voltage command value V _γδ corresponding to the calculation result is generated by the voltage command value generation unit 19B, and this is converted into a three-phase voltage command value V _uvw through the coordinate conversion units 20A and 20B. The drive circuit 13 operates with a duty ratio corresponding to the three-phase voltage command value V _uvw by the action of the PWM control unit 21, so that the motor 1 is driven and an assist torque corresponding to the target current value I _dq is generated. The steering mechanism 3 will be given. Thus, steering assistance can be performed according to the operation torque and the vehicle speed. The three-phase detection current I _uvw detected by the current detection unit 11 passes through the coordinate conversion units 17A and 17B, and is expressed by a two-phase rotational coordinate system (γ−δ) so as to correspond to the target current value I _dq. After being converted to the phase detection current I _γδ , it is given to the deviation calculating section 18.

In order to perform coordinate conversion between the rotating coordinate system and the fixed coordinate system, the rotation angle (phase angle, ie, electrical angle) θ of the rotor 50 is required. The control rotation angle θ ^ representing the rotation angle is generated by the rotation angle calculation unit 22 using the output of the resolver 2 or is estimated by the estimation calculation by the rotation angle estimation unit 31. The control rotation angle θ ^ calculated by either of them is given from the switching unit 32 to the αβ / γδ coordinate conversion unit 17B and the γδ / αβ coordinate conversion unit 20A. Hereinafter, the rotation angle detected by the rotation angle calculation unit 22 is referred to as “detected rotation angle θ _S ”, and the rotation angle estimated by the rotation angle estimation unit 31 is referred to as “estimated rotation angle θ _E ”.

The rotation angular velocity calculation unit 23 divides the difference Δθ _S of the detected rotation angle θ _S given every predetermined control cycle (for example, 200 μsec) from the rotation angle calculation unit 22 by the control cycle, thereby calculating the rotation angular velocity ω of the rotor 50. Calculate.
The sensor failure determination unit 25 determines whether or not there is a failure in the resolver 2. For example, the sensor failure determination unit 25 can detect a failure of the resolver 2, a disconnection failure of the signal line 2a, and a grounding failure of the signal line 2a by monitoring a signal derived to the signal line 2a of the resolver 2. . More specifically, the signal line 2a between the resolver 2 and the motor control device 10 is connected to the power supply potential via a pull-up resistor or connected to the ground potential via a pull-down resistor. Can do. In this case, when the signal line 2a is disconnected, the signal (sine signal or cosine signal) from the resolver 2 is not derived to the signal line 2a. Instead, the signal line 2a is fixed to the power supply potential or the ground potential. Is done. Therefore, the sensor failure determination unit 25 can determine whether or not the resolver 2 has failed (including signal line failure) by determining whether or not the signal line 2a is fixed at the power supply potential or the ground potential. . Of course, other known methods may be applied to the failure detection of the resolver 2.

The initial value determination unit 24 determines an initial value for the estimation calculation in the rotation angle estimation unit 31. In response to a trigger from the sensor failure determination unit 25, the initial value determination unit 24 performs an operation for determining the initial value. More specifically, the initial value determination unit 24 is detected by the rotation angular velocity ω calculated by the rotation angular velocity calculation unit 23, the two-phase voltage command value V _αβ , the two-phase detection current I _αβ, and the torque sensor 7. The initial value is obtained based on the operating torque. Since the operating torque changes according to the torque generated by the motor 1, the output signal of the torque sensor 7 corresponds to the generated torque of the motor 1. That is, in this embodiment, the torque sensor 7 not only detects the operation torque applied to the steering wheel, but also functions as a torque detection means for detecting the torque generated by the motor 1.

The search voltage generator 26 generates a search voltage command value for generating a search magnetic field from the stator 55 for roughly specifying the direction of the rotor 50 when the resolver 2 fails, and superimposes it on the two-phase voltage command value V _αβ . To do. This exploration voltage generator 26 operates in response to a trigger from the sensor failure determination unit 25. When the exploration magnetic field is generated, torque corresponding to the exploration magnetic field is generated in the rotor 50. This torque is detected by the torque sensor 7 and given to the initial value determination unit 24.

The switching unit 32 selects the detected rotation angle θ _S or the estimated rotation angle θ _E according to the determination result by the sensor failure determination unit 25, and outputs the selected rotation angle as the control rotation angle θ ^. That is, at the normal time when the sensor failure determination unit 25 determines that the resolver 2 has not failed, the switching unit 32 outputs the detected rotation angle θ _S as the control rotation angle θ ^. On the other hand, when the sensor failure determination unit 25 determines that a failure has occurred in the resolver 2, that is, when the resolver 2 has failed, the estimated rotation angle θ _E is selected and output as the control rotation angle θ ^. As a result, the motor 1 is driven by a sensorless driving method that does not use the resolver 2 as a rotation angle sensor. That is, even when the resolver 2 fails, the driving of the motor 1 can be continued and the steering assist force can be applied to the steering mechanism 3.

The rotation angle estimation unit 31 is based on the motor current (the two-phase detection current I _αβ converted by the UVW / αβ coordinate conversion unit 17A) and the two-phase voltage command value V _αβ given from the γδ / αβ coordinate conversion unit 20A. Thus, the estimated rotation angle θ _E is calculated. The rotation angle estimator 31 starts a calculation process upon receiving a trigger from the sensor failure determiner 25. That is, during the normal time when the sensor failure determination unit 25 determines that no failure has occurred in the resolver 2, the estimation calculation for obtaining the estimated rotation angle θ _E is not performed. When the sensor failure determination unit 25 determines that a failure has occurred in the resolver 2, the estimation calculation is started using the initial value given from the initial value determination unit 31, and the estimated rotation angle θ _E is generated and switched. Part 32 is given.

The switch 30 is controlled to an on state in which the two-phase voltage command value V _αβ and the two-phase detection current I _αβ are given to the rotation angle estimation unit 31 and an off state in which they are not given to the rotation angle estimation unit 31. The switch 30 has a switch function by software processing. The switch 30 is controlled to be turned on / off according to the determination result by the sensor failure determination unit 25. That is, the switch 30 is controlled to be in an off state at the normal time when the sensor failure determination unit 25 determines that the resolver 2 has not failed. On the other hand, when the sensor failure determination unit 25 determines that a failure has occurred in the resolver 2, the sensor failure determination unit 25 is controlled to be in an on state. Therefore, the two-phase voltage command value V _αβ and the two-phase detection current I _αβ are given to the rotation angle estimation unit 31 only when a failure occurs in the resolver 2.

FIG. 3 is a block diagram illustrating a configuration example of the rotation angle estimation unit 31. The rotation angle estimation unit 31 includes an induced voltage observer 35 serving as an induced voltage estimation unit, and a rotor angle estimation unit 36 serving as a conversion unit that converts the estimated induced voltage into a rotor rotation angle. The induced voltage observer 35 calculates the induced voltage of the motor 1 based on the two-phase voltage command value V _αβ from the γδ / αβ coordinate conversion unit 20A and the two-phase detection current I _αβ from the UVW / αβ coordinate conversion unit 17A. Then, an α-axis induced voltage estimated value E ^ _α and a β-axis induced voltage estimated value E ^ _β (hereinafter collectively referred to as “estimated induced voltage E ^ _αβ ”) are generated. The rotor angle estimation unit 36 estimates the rotor rotation angle based on the estimated induced voltage E ^ _αβ , and generates the estimated rotation angle θ _E.

FIG. 4 is a block diagram for explaining a configuration example of the induced voltage observer 35. As the induced voltage observer 35, for example, the configuration disclosed in Non-Patent Document 1 can be used, and more specifically, the configuration is as shown in FIG.
The induced voltage observer 35 is expressed by the following equations (3) and (4).
x _αβ (n + 1) = A · x _αβ (n) + B1 · V _αβ (n) + B2 · I _αβ (n) (3)
E ^ _αβ (n) = C · x _αβ (n) + D1 · V _αβ (n) + D2 · I _αβ (n) (4)
x _αβ is a variable (internal variable) representing an internal state, and includes an α-axis variable x _α corresponding to the α-axis and a β-axis variable x _β corresponding to the β-axis. A, B1, B2, C, D1, and D2 are matrix operators of two rows and two columns, and are configured by motor parameters (resistance and inductance) and observer gain, or “0”. n represents a calculation cycle number (that is, time of sampling timing), and the calculation cycle (time) n is a calculation cycle immediately before the calculation cycle (time) n + 1. Thus, for example, x _αβ (n) represents an internal variable at time n.

The initial value determination unit 24 (see FIG. 1) determines an initial value of the internal variable x _αβ , and an induced voltage estimation calculation by the induced voltage observer 35 is performed based on the initial value. If this initial value is appropriate, the estimated induced voltage E ^ _αβ quickly converges to an appropriate value.
On the other hand, the induced voltage E _αβ can be expressed by the following equation (5). Where K _E is an induced voltage constant, θ is a rotor rotational position, and ω is a rotor rotational angular velocity.

したがって、推定誘起電圧Ｅ^_αβが求まれば、次の(6)式に従って、推定回転角θ_Eが求まる。この演算が、ロータ角推定部３６によって行われるようになっている。

Therefore, if the estimated induced voltage E ^ _αβ is obtained, the estimated rotation angle θ _E is obtained according to the following equation (6). This calculation is performed by the rotor angle estimation unit 36.

図５は、探査電圧発生部２６が発生する探査電圧指令値によってステータ５５が形成する探査磁界を説明するための図である。ステータ巻線５１〜５３に探査電圧指令値に相当する探査電圧が印加されることによって、ステータ５５側の電磁石による仮想的なＮ極が、図５に示す所定の複数個（この実施形態では１２個）の位置Ｐ１〜Ｐ１２に、この順序で、制御周期（たとえば、２００μｓｅｃ）毎に順次形成される。たとえば、二相固定座標系（α−β）におけるα軸からの角度θ_Fによって仮想的なＮ極の位置を表すとすると、図５に示す各位置Ｐ１，Ｐ２，…，Ｐ１２は、次のように表される。

FIG. 5 is a diagram for explaining the exploration magnetic field formed by the stator 55 by the exploration voltage command value generated by the exploration voltage generator 26. By applying an exploration voltage corresponding to the exploration voltage command value to the stator windings 51 to 53, there are a predetermined number of virtual N poles by the electromagnet on the stator 55 side (12 in this embodiment). ) Positions P1 to P12 are sequentially formed in this order every control period (for example, 200 μsec). For example, if the virtual N-pole position is represented by the angle θ _F from the α axis in the two-phase fixed coordinate system (α−β), the positions P1, P2,..., P12 shown in FIG. It is expressed as follows.

P1: Angle θ _F = 90 °
P2: Angle θ _F = 270 ° (P1 angle + 180 °)
P3: Angle θ _F = 60 ° (P2 angle + 150 °)
P4: Angle θ _F = 240 ° (P3 angle + 180 °)
P5: Angle θ _F = 30 ° (P4 angle + 150 °)
P6: Angle θ _F = 210 ° (P5 angle + 180 °)
P7: Angle θ _F = 0 ° (P6 angle + 150 °)
P8: Angle θ _F = 180 ° (P7 angle + 180 °)
P9: Angle θ _F = 330 ° (P8 angle + 150 °)
P10: Angle θ _F = 150 ° (P9 angle + 180 °)
P11: Angle θ _F = 300 ° (P10 angle + 150 °)
P12: Angle θ _F = 120 ° (P11 angle + 180 °)
That is, the positions P1 to P12 are set to positions obtained by dividing the entire 360 ° angle range at equal intervals of 30 °. The positions where the virtual N poles are formed before and after each other are positions facing each other with respect to the rotation center of the rotor 50. More specifically, the position where the virtual N pole is formed before and after is 180 degrees apart or 180 ° −Δ (Δ is a predetermined angle, and in the above example, Δ = 30 °. ). More specifically, a first step of forming a virtual N pole at a position 180 ° away from the immediately preceding virtual N pole formation position, and 180 ° −Δ (Δ = 30) from the virtual N pole position formed in the first step. By alternately executing the second step of forming the virtual N pole at a position separated by 150 ° at the time of °, the virtual N pole position is sequentially changed.

Thus, the exploration magnetic field is applied one after another while changing its direction for each control period, and the order of the directions is selected so that the phase is substantially reversed. Thereby, the search magnetic field of the several direction over the whole angle range of a rotor rotation angle (electrical angle) can be formed sequentially, suppressing rotation of the rotor 50 by a search magnetic field.
The output signal of the torque sensor 7 when the exploration magnetic field in each direction is formed is taken into the initial value determination unit 24 and used to determine the aforementioned initial value.

FIG. 6 is a diagram for explaining the relationship between the direction of the search magnetic field and the torque generated in the rotor 50 by the search magnetic field. The torque generated in the rotor 50 is detected by the torque sensor 7 as described above.
In FIG. 6, assuming that the rotor 50 is positioned at the rotation angle shown in FIG. 5 (when the N pole of the rotor 50 faces the position P5), the positions of the virtual N poles P1 to P12 are assumed. The relationship between (ie, the direction of the exploration magnetic field) and the torque generated in the rotor 50 is shown. Further, the counterclockwise torque (clockwise in the αβ coordinate plane) generated in the rotor 50 is represented by a positive value, and the clockwise torque generated in the rotor 50 (clockwise in the αβ coordinate plane) is represented by a negative value.

  When the search magnetic field is orthogonal to the magnetic pole direction of the rotor 50, the largest torque is generated. That is, when there is a virtual N pole on the stator 55 side at a position of −90 ° with respect to the N pole position of the rotor 50 (position P11), the torque of the rotor 50 takes a maximum value (positive value). When there is a virtual N pole on the stator 55 side at a position of + 90 ° with respect to the N pole position (position P12), the torque of the rotor 50 takes a minimum value (negative value).

When the exploration magnetic field is parallel to the magnetic pole direction of the rotor 50, the magnitude of torque generated in the rotor 50 is minimized. That is, when there is a virtual N pole on the stator 55 side at a position of 0 ° or 180 ° with respect to the N pole position of the rotor 50 (positions P5 and P6), the torque of the rotor 50 becomes zero.
FIG. 7 is a flowchart for explaining the initial value determining process by the initial value determining unit 24. When the exploration voltage generator 26 generates the exploration voltage (step S21), the initial value determination unit 24 operates the direction of the exploration magnetic field (virtual N pole position on the stator 55 side) and the torque sensor 7 to output. Torque (generated torque of the motor 1) is collected in association with each other (step S22). As a result, a torque graph as shown in FIG. 6 is obtained. From this graph, the initial value determination unit 24 obtains at least one of the zero cross points Z1, Z2, the maximum point Lmax, and the minimum point Lmin of the detected torque (step S23). Based on this, the initial value determination unit 24 obtains the rotation angle θ (n) (0 ° ≦ θ (n) <360 °) of the rotor 50 (step S24).

For example, if the zero cross point Z1 is obtained, it can be seen from the change in the detected torque near the zero cross point Z1 that the zero cross point Z1 changes from a negative value to a positive value. Therefore, the initial value determination unit 24 specifies the zero cross point Z1 (position P5) as the rotation angle θ (n) of the rotor 50.
If the zero cross point Z2 is obtained, it can be seen from the change in the detected torque near the zero cross point Z2 that the zero cross point Z2 changes from a positive value to a negative value. Therefore, the initial value determination unit 24 specifies the position obtained by adding or subtracting 180 ° to the zero cross point Z2 (position P6) as the rotation angle θ (n) of the rotor 50.

If the maximum point Lmax is to be obtained, the initial value determination unit 24 specifies the position obtained by adding 90 ° or subtracting 270 ° from the maximum point Lmax (position P11) as the rotation angle θ (n) of the rotor 50. To do.
If the minimum point Lmin is to be obtained, the initial value determination unit 24 specifies a position obtained by subtracting 90 ° from the minimum point Lmin (position P12) or adding 270 ° as the rotation angle θ (n) of the rotor 50.

When the rotation angle θ (n) of the rotor 50 is specified in this way, the initial value determination unit 24 substitutes the rotation angle θ (n) into the equation (5) to thereby generate the induced voltage E _αβ (n). Obtained (step S25). At this time, the rotational angular velocity ω is required. As the rotation angular velocity ω, a value calculated by the rotation angular velocity calculation unit 23 just before the sensor failure determination unit 25 determines the occurrence of the failure may be used. The induced voltage constant K _E is it is sufficient to measure in advance.

Next, the initial value determination unit 24 obtains an internal variable x _αβ (n) according to the following equation (7) obtained by transforming the equation (4), and uses this as the initial value of the internal variable x _αβ (n). (Step S26).
x _αβ (n) = C ⁻¹ {E ^ _αβ (n) −D1 · V _αβ (n) + D2 · I _αβ (n)} (7)
In order for this calculation to be possible, the matrix operator C needs to be regular (having an inverse matrix). In general, the matrix operator C is regular because it is a diagonal matrix or a unit matrix.

  In the case where virtual N poles are formed at 12 positions when the control cycle is 200 μsec (that is, when the exploration magnetic field is formed over 12 times), the required time is 2.4 msec. In this short time, the external torque, that is, the fluctuation of the operation torque applied to the steering wheel by the driver can be ignored. Therefore, the torque response of the motor 1 can be detected by using the output signal from the torque sensor 7 during this period.

In addition, since the rotation speed of the steering wheel has an upper limit value (the maximum rotation speed that can be operated by the driver), the rotation speed of the rotor 50 has an upper limit value accordingly. Even when the rotor 50 rotates at the upper limit value, for example, the rotation angle of the rotor 50 within a time of 2.4 msec is at most about 45 °. Therefore, the error of the rotation angle of the rotor 50 obtained as described above is 45 ° at most. If the initial value of the internal variable x _αβ (n) is determined within such an error range, the estimated rotation angle θ _E can be promptly set to an appropriate value (after the estimation calculation by the rotation angle estimation unit 31 is started). The value converges to a value approximating the actual rotation angle of the rotor 50.

FIG. 8 is a flowchart for explaining processing that the microcomputer 12 repeatedly executes at predetermined control cycles. The microcomputer 12 takes in the output signals of the torque sensor 7, the vehicle speed sensor 8, the resolver 2, and the current detector 11 (step S1). The basic target current value calculation unit 15 calculates the basic target current value I ^* based on the operation torque detected by the torque sensor 7 and the vehicle speed detected by the vehicle speed sensor 8 (step S2). On the other hand, the sensor failure determination unit 25 determines the presence or absence of a failure of the resolver 2 based on a signal derived to the signal line 2a (step S3).

If there is no failure in the resolver 2 (step S3: NO), the switching unit 32 selects the detected rotation angle θ _S (rotation angle by the resolver output) calculated by the rotation angle calculation unit 22 as the control rotation angle θ ^ ( Step S4). Thus, normal control for driving the motor 1 is executed based on the basic target current value I ^* using the output signal of the resolver 2 (step S7). More specifically, the d-axis target current value I _d ^* and the q-axis target current value I _q ^* are set by the dq-axis target current value calculation unit 16. Also, the three-phase detection current I _uvw detected by the current detection unit 11 is coordinate-converted by the coordinate conversion units 17A and 17B, and the γ-axis current I _γ and the δ-axis current I _δ are obtained. The deviation calculation unit 18 calculates a d-axis current deviation δI _d (= I _d ^* −I _γ ) and a q-axis current deviation δI _q (= I _q ^* −I _δ ). The PI control unit 19A performs a PI (proportional integration) calculation or the like on the current deviations δI _d and δI _q , and based on this PI calculation, the voltage command value generation unit 19B performs the d-axis voltage command value V _d ^* and the q-axis. A voltage command value V _q ^* is generated. These are subjected to coordinate conversion by the coordinate conversion units 20A and 20B, thereby generating UVW phase voltage command values V _u ^* , V _v ^* and V _w ^* . A PWM control signal corresponding to these voltage command values V _u ^* , V _v ^* , V _w ^* is generated by the PWM control unit 21. In the coordinate conversion calculation in the αβ / γδ coordinate conversion unit 17B and the γδ / αβ coordinate conversion unit 20A, the detected rotation angle θ _S calculated by the rotation angle calculation unit 22 based on the output signal of the resolver 2 is the control rotation angle θ ^. (Step S4).

When this normal control is executed, the switch 30 is turned off, and the internal variable x _αβ (n) of the rotation angle estimation unit 31 is reset (step S5). Further, the rotation angle estimation calculation by the rotation angle estimation unit 31 is stopped (step S6). Thereby, since the calculation load is suppressed, it is not always necessary to apply an expensive computer 12 with high calculation capability.

On the other hand, the sensor failure determining unit 25, a fault in the resolver 2 is determined to have occurred (step S3: YES), the switching unit 32, the estimated rotation angle theta _E of the rotational angle estimation unit 31 calculates the control rotation The angle θ ^ is selected (step S8). Then, the switch 30 is controlled to be in an ON state (step S9), and the two-phase voltage command value V _αβ and the two-phase detection current I _αβ are supplied to the rotation angle estimation unit 31.

Further, it is determined whether or not the initial value of the internal variable x _αβ (n) has been determined (step S10). If the initial value is not yet determined (if the internal variable x _αβ (n) is in the reset state), the initial value is determined. Initial value determination processing (see FIG. 7) is performed by the function of the value determination unit 24 and the like. That is, a search voltage command value for generating a search magnetic field is generated from the search voltage generator 26 (step S11). Then, initial value determination is performed by the initial value determination unit 24 (step S12). Details of this processing are as described above.

When the initial value of the internal variable x _αβ (n) is determined in this way, the position estimation calculation is performed by the rotation angle estimation unit 31 using this initial value (step S13). If the initial value has been determined (step S10: YES), that is, if the rotation angle estimation calculation by the rotation angle estimation unit 31 has already been started, the processing of steps S11 and S12 is omitted.
Then, the motor 1 is driven by sensorless control using the estimated rotation angle θ _E obtained by the rotation angle estimation unit 31 (step S7).

Thus, in this embodiment, when a fault in the resolver 2 is generated, using the output signal of the sensor-less control (resolver 2 using the estimated rotational angle theta _E obtained by the rotational angle estimation unit 31 as a control rotational angle theta ^ Without motor control). Thereby, even after the resolver 2 breaks down, the drive of the motor 1 can be continued and the steering assist force can be applied to the steering mechanism 3.

  In normal times when the resolver 2 has not failed, the estimation calculation by the rotation angle estimation unit 31 is in a stopped state, so the calculation load is reduced. For this reason, an inexpensive computer 12 having a low computing capacity can be applied to the microcomputer 12 as compared with the case of the configuration in which the estimation calculation is performed even during normal times. Thereby, the cost reduction of the motor control apparatus 10 and the cost reduction of an electric power steering apparatus can be aimed at by extension.

Further, when a failure occurs in the resolver 2, the rotation angle of the rotor 50 is roughly determined by the formation of the exploration magnetic field. Based on this, the initial value of the internal variable x _αβ (n) in the rotation angle estimation unit 31 is determined. Is done. As a result, the rotation angle estimation unit 31 can quickly generate a reasonable estimated rotation angle θ _E after the start of the estimation calculation. Thereby, sensorless control using the estimated rotation angle θ _E can be quickly started up, and the motor 1 can be appropriately controlled using the estimated rotation angle θ _E with high accuracy. Therefore, since the motor 1 can be driven efficiently, a sufficient torque can be generated.

As mentioned above, although one Embodiment of this invention was described, this invention can also be implemented with another form. For example, in the above-described embodiment, the rotation angular velocity ω is calculated using the detected rotation angle θ _S calculated by the rotation angle calculation unit 22. However, as indicated by a two-dot chain line 60 in FIG. When the steering angle sensor 9 for detecting the steering angle is provided, the rotational angular velocity ω of the motor 1 may be obtained from the steering angle detected by the steering angle sensor 9.

Further, in the above-described embodiment, the torque generated by the rotor 50 when the exploration magnetic field is formed is detected by the torque sensor 7 for detecting the operation torque, but for detecting the torque of the motor 1. A torque sensor may be provided separately.
In the above-described embodiment, the virtual N poles are formed at 12 positions by the exploration magnetic field, but the number of virtual N poles formed is not limited to 12.

In the above-described embodiment, the case where the present invention is applied to control the motor 1 as a drive source of the electric power steering apparatus has been described. However, the present invention is applicable to motor control for applications other than the electric power steering apparatus. It can also be applied to.
In addition, various design changes can be made within the scope of matters described in the claims.

1 is a block diagram for explaining an electrical configuration of an electric power steering apparatus to which a motor control device according to an embodiment of the present invention is applied. FIG. It is a figure for demonstrating the structure and coordinate system of a motor. It is a block diagram which shows the structural example of a rotation angle estimation part. It is a block diagram for demonstrating the structural example of an induced voltage observer. It is a figure for demonstrating the search magnetic field which a stator forms with the search voltage command value which a search voltage generation part produces | generates. It is a figure for demonstrating the relationship between the direction of a search magnetic field, and the torque which arises in a rotor by a search magnetic field. It is a flowchart for demonstrating the initial value determination process by the initial value determination part. It is a flowchart for demonstrating the process which a microcomputer repeatedly performs for every predetermined control period.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Motor control apparatus, 11 ... Current detection part, 12 ... Microcomputer, 50 ... Rotor, 51-53 ... Stator winding, 55 ... Stator

Claims (3)

A motor control device for controlling a motor including a rotor and a stator facing the rotor,
Rotation angle detection means for detecting the rotation angle of the rotor;
A rotation angle estimating means for estimating a rotation angle of the rotor;
Torque detecting means for detecting torque generated by the motor;
A search magnetic field generation means for generating a search magnetic field from the stator when the rotation angle detection means fails;
An initial value determining means for determining an initial value for an estimation calculation in the rotation angle estimating means based on a detected torque detected by the torque detecting means when the exploration magnetic field is generated;
The motor is controlled based on the detected rotation angle detected by the rotation angle detection means at a normal time when the rotation angle detection means has not failed. When the rotation angle detection means fails, the rotation angle estimation means sets the initial value. And a control means for controlling the motor based on the estimated rotation angle estimated on the basis of the motor control device. The exploration magnetic field generating means sequentially generates a plurality of exploration magnetic fields in different directions,
The motor control device according to claim 1, wherein the initial value determining means determines the initial value based on a direction of a search magnetic field corresponding to a zero cross point of a detected torque. The exploration magnetic field generating means sequentially generates a plurality of exploration magnetic fields in different directions,
The motor control device according to claim 1, wherein the initial value determining unit determines the initial value based on a direction of a search magnetic field corresponding to a maximum point or a minimum point of the detected torque.

Images