JP7501190B2 - Magnetic pole position estimation device and control device for AC synchronous motor - Google Patents

Description

The present invention relates to a magnetic pole position estimation device for an AC synchronous motor, such as a linear motor or a rotary direct drive (DD) motor, that estimates the actual magnetic pole position by passing a predetermined pull-in current through the motor in order to eliminate the error (phase difference) between the actual magnetic pole position (hereinafter also referred to as the real axis) of the AC synchronous motor and the magnetic pole position estimated for control (hereinafter also referred to as the estimated axis), and a control device that controls the AC synchronous motor using the estimated magnetic pole position.

As a conventional technique of this kind, for example, a magnetic pole position estimation method for an AC synchronous motor is known, as described in Japanese Patent Application Laid-Open No. 2003-233696.
This conventional technology performs position/speed control including proportional-integral control by comparing a reference magnetic pole position command given to the motor with a magnetic pole position detection value, and updates the controlled magnetic pole position by adding a magnetic pole position error estimate corresponding to the positive or negative moving direction of the motor's rotor (or mover) to the motor's electrical angle position, thereby controlling the motor in a direction that reduces the error torque while changing the application direction of the estimation current signal.Then, the magnetic pole position error estimate value at the point when the error torque finally becomes zero and the motor stops is regarded as a true value, and the actual magnetic pole position is estimated to control the motor.
This prevents erroneous estimation when the magnetic pole position is shifted by 180 degrees, and enables estimation of the magnetic pole position without being affected by static friction.

JP 2009-183022 A ([0017] to [0019], Figs. 1 to 3, etc.)

In the conventional technology described above, a magnetic pole position error estimate is obtained by operating the position/speed control unit so that the magnetic pole position detection value coincides with the magnetic pole position command, and this magnetic pole position error estimate is then added to an estimated value of the motor's electrical angle to perform coordinate transformation, thereby generating a voltage command for the inverter that drives the motor.
However, when the inertia of the motor changes significantly, the gain adjustment time required to set the position/speed control gain to the optimum value becomes long. Also, when driving a DD motor, the inertia ratio of the motor can be several hundred to several thousand times, so the band of the position/speed control gain must be set wide, resulting in the problem of unstable operation.

The problem to be solved by the present invention is to provide a magnetic pole position estimation device for an AC synchronous motor that can estimate the magnetic pole position in a short time regardless of the magnitude of the motor's inertia, and that reduces the amount of movement of the motor's rotor and mover during the estimation operation so as not to damage the load, and a control device for an AC synchronous motor that uses this magnetic pole position estimation device.

In order to solve the above problem, a magnetic pole position estimating device for an AC synchronous motor according to claim 1 is a magnetic pole position estimating device that estimates an actual magnetic pole position of the AC synchronous motor based on a movement amount electrical angle obtained by converting a movement amount of a rotor or a mover of the AC synchronous motor when a predetermined current is caused to flow through an armature winding of the AC synchronous motor based on a current command, the magnetic pole position estimating device comprising:
The control phase angle obtained by multiplying the movement amount electrical angle by a predetermined variable gain is used to perform coordinate transformation for feeding back a current flowing through the armature winding , and an electrical angle phase offset value to be added to the movement amount during normal operation of the AC synchronous motor is calculated.

The magnetic pole position estimation device for an AC synchronous motor according to claim 2 is the magnetic pole position estimation device for an AC synchronous motor described in claim 1, characterized in that the variable gain is a gain that decreases over time from a preset initial value to a lower limit value of 1.

The magnetic pole position estimation device of an AC synchronous motor according to claim 3 is the magnetic pole position estimation device of an AC synchronous motor described in claim 1 or 2, characterized in that an alarm is generated and the magnetic pole position estimation operation is stopped when a timeout time has elapsed since the specified current was caused to flow.

The magnetic pole position estimation device for an AC synchronous motor according to claim 4 is the magnetic pole position estimation device for an AC synchronous motor according to any one of claims 1 to 4, characterized in that, if the amount of movement does not exceed a predetermined value, an alarm is generated and the magnetic pole position estimation operation is stopped.

The control device for an AC synchronous motor according to claim 5 is characterized in that the electrical angle phase offset value is calculated as a magnetic pole position error by the magnetic pole position estimation device according to any one of claims 1 to 4, and during normal operation of the AC synchronous motor, coordinate conversion for current control and voltage control is performed using a control phase angle obtained by adding the electrical angle phase offset value to the movement amount.

According to the present invention, when estimating the magnetic pole position, the coordinate transformation phase is directly manipulated without performing position/speed control as in the prior art. Therefore, even when the inertia of the motor changes significantly or the inertia ratio is large, there are no problems such as a long adjustment time for the position/speed control gains or unstable adjustment.
That is, for low inertia loads, the magnetic pole position estimation time is shortened, and the amount of movement of the rotor, etc. during the estimation operation is reduced while preventing sudden operation, and for large inertia loads, a timeout alarm can be generated as necessary to prevent erroneous estimation when the amount of motor movement does not converge to zero due to the pull-in operation and becomes oscillatory, or when the motor movement amount is in an oscillatory state and a stop decision is mistakenly made near zero.

FIG. 2 is a block diagram showing a configuration of a control device according to the embodiment of the present invention. 5 is a flowchart showing a magnetic pole position estimation operation in the embodiment of the present invention. FIG. 3 is a phasor diagram for sequence SQ1 of FIG. 2. FIG. 3 is a phasor diagram for sequence SQ2 of FIG. 2. FIG. 3 is a phasor diagram at the end of sequence SQ3 in FIG. 2. FIG. 4 is a waveform diagram showing a magnetic pole position estimation operation in the embodiment of the present invention.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing the configuration of a control device according to this embodiment, and shows a configuration for attracting the magnetic poles of an AC synchronous motor (hereinafter simply referred to as a motor) 50 such as a linear motor or a DD motor, to estimate the magnetic pole position.

1, a pull-in current command generating unit 10 outputs a d-axis current command i _d ^* to pass a d-axis current through the armature coil of a motor 50 when a magnetic pole is pulled in by a magnetic pole position estimation operation. As is well known, the d-axis is a coordinate axis along the magnetic pole direction of the motor 50, and the coordinate axis perpendicular to this d-axis is the q-axis.
The d-axis current command i _d ^* is input to the current control unit 20 together with the d-axis current i _d output from a uvw/dq coordinate conversion unit 32 described later. The current control unit 20 generates the d-axis voltage command v d * and the q-axis voltage command v q * based on the d-axis current command i _d ^* and the q-axis current command i _q ^* and the fed-back detection values (calculated values) i _d and i _q not ^only when estimating the magnetic pole position but also during normal operation by vector control _{or the like, but the q-axis current command i q *} ^and the q-axis voltage command v _q ^* are not necessary when estimating the magnetic pole position by ^the magnetic _pole entrainment method.

The current control unit 20 performs an adjustment calculation so that the d-axis current i _d coincides with the d-axis current command i _d ^* to generate a d-axis voltage command v _d ^* . This d-axis voltage command v _d ^* is input to a dq/uvw coordinate conversion unit 31 together with a q-axis voltage command v _q ^* (= 0), and the dq/uvw coordinate conversion unit 31 converts the d-axis and q-axis voltage commands v _{d *} , v _q ^* in the orthogonal rotating coordinate system into three-phase voltage commands v _u ^* , v _v ^* , v _w ^* in the stationary coordinate system using a control phase angle θ ^ec as an estimated electrical angle.

The voltage commands _vu ^* , _vv ^* , _vw ^* are input to a PWM calculation unit 40, which controls internal semiconductor switching elements to be turned on and off by pulse width modulation control in accordance with the voltage commands _vu ^* , _vv ^* , _vw ^* , and supplies three-phase currents _iu , _iv , _iw to the motor 50. A pulse generator (PG) 60 is attached to the motor 50 for detecting the mechanical angle _θm of the rotor or mover.

The movement amount mechanical angle _θm output from the pulse generator 60 is converted to a movement amount electrical angle _θe (= _θm /P, P: number of pole pairs of the motor 50) by an electrical angle calculation unit 70 and output. This movement amount electrical angle _θe is multiplied by a variable gain _Kth and output as the above-mentioned control phase angle _θec (= _θe × _Kth ) to each of the coordinate conversion units 31 and 32. Note that the variable gain _Kth is a gain whose value decreases by 1 every 10 [ms] from a preset arbitrary initial value _K0 to a minimum value of 1, for example.

Of the three-phase currents _iu , _iv , and _iw supplied to the motor 50, for example, two-phase currents _iu and _iw are detected by a current detection unit 80 and input to the uvw/ _dq coordinate conversion unit 32. Here, the remaining one-phase current _iv is found by calculation ( _iv = -iu _-iw ), but it goes without saying that the two-phase currents detected by the current detection unit 80 may be other combinations.
The uvw/dq coordinate conversion unit 32 converts the three-phase currents _iu , _iv , and _iw into a two-phase d-axis current _id and a q-axis current _iq using the control phase angle _θec , and outputs them to the current control unit 20.

Next, the operation of estimating the magnetic pole position in this embodiment will be described with reference to the flowchart in Fig. 2, the phasor diagrams in Fig. 3 to Fig. 5, and the waveform diagram in Fig. 6. Note that in this embodiment, a DD motor is used as the motor 50.
2, first, in sequence SQ1, an initial pull-in operation is executed (step S1). In this initial pull-in operation, the amount of movement of the real axis when the d-axis current i _d is applied to the motor 50 in accordance with the d-axis current command i _d ^* is fed back to the control phase angle so as to make the magnetic pole position error, i.e., the phase difference Δθ between the real axis and the estimated axis, zero.

The left side of Figure 3 shows the relationship between the actual axis and the estimated axis at the start of the initial retraction operation, and the time elapsed since the start of the retraction operation is counted by a retraction operation counter (not shown). The operation of this retraction operation counter is shown in the upper part of Figure 6. The retraction operation counter is reset at the same time that the motor stop counter, which will be described later, is reset, but if the count value of the retraction operation counter is not reset and continues to increase beyond a specified timeout time (retraction current continues to flow), it is determined that an abnormality has occurred, an alarm is generated, and the retraction operation is stopped.

As shown in FIG. 6, when the pulling operation starts, the rotor movement amount electrical angle _θe increases, for example, in the negative direction, and at the same time, the variable gain K _th decreases by 1 every 10 ms from the initial value K ₀ .
As a result, the control phase angle _θec increases rapidly at the start of the pull-in operation, but starts to decrease when the variable gain _Kth becomes 1, and then maintains an almost constant positive value. When the rotor movement amount electrical angle _θe becomes | _C1 | or less, it is determined for the sake of convenience that this is a "movement amount zero" state, and from that point on, the motor stop counter starts counting.

When the time corresponding to the count value of the motor stop counter reaches the set time _Ts , it is determined that the initial pull-in operation is completed, power supply to the motor 50 is stopped (Yes in step S2 of FIG. 2), and the d-axis current command _id ^* is decreased to 0 over a time _Tv .
Here, the left side of Fig. 3 is a phasor diagram at the start of sequence SQ1, and the right side is a phasor diagram at the end of sequence SQ1. When the initial value K ₀ of variable gain K _th is 1, there is a relationship between the actual axis movement amount θ _a1 calculated from the phase angle calculated by electrical angle calculation unit 70 in Fig. 1 and its previous value and the movement amount (estimated axis movement amount) θ _e1 of the control phase angle, θ _e1 = -θ _a1 = Δθ/2. Here, Δθ is a phase difference equivalent to the magnetic pole position error.
This completes the initial pull-in operation, and the process proceeds to sequence SQ2 (step S3 in FIG. 2).

This sequence SQ2 is an operation for preventing 180 deg inversion estimation.
The 180° reversal estimation is a process for preventing erroneous estimation when the magnetic pole position is shifted by 180°. In other words, if the magnetic pole position error is close to 180°, no error torque is generated in the motor, so the magnetic pole position cannot be estimated, and it is not possible to determine whether the magnetic pole position error is 0 or 180°. For this reason, by shifting the control phase angle _θec by 90° and performing a pull-in operation, a current is passed through the motor 50 in the q-axis direction, creating a state in which the rotor always moves.

In particular, in a linear motor, since the movable range of the mover is limited, if the retraction operation is performed in only one direction, the mover may reach the end and the retraction operation may not be performed correctly. Therefore, in sequence SQ2, the sign is determined so that the mover moves in the opposite direction to sequence SQ1, and the control phase angle _θec is manipulated.

The initial control phase angle in sequence SQ2 is adjusted as follows so that it is in the opposite direction to the real axis movement direction (rotor movement direction) in sequence SQ1.
(1) When the real axis movement direction in sequence SQ1 is positive:
The amount of movement of the initial control phase angle in the sequence SQ2 is θ _e1 −90 deg.
(2) When the real axis movement direction in sequence SQ1 is negative:
The amount of movement of the initial control phase angle in the sequence SQ2 is θ _e1 +90 deg.
In Figures 3 and 6, since the real axis movement direction in sequence SQ1 is negative, in sequence SQ2, as shown in the phasor diagram on the left side of Figure 4, a current is passed to the motor 50 in the q-axis direction based on the d-axis current command i _d ^* in a state where the control phase angle θ _ec has been rotated by 90 degrees, and the rotor is moved in the opposite direction to that of sequence SQ1.

As shown in Fig. 6, in sequence SQ2, the magnitude of the d-axis current command i _d ^* is the same as in sequence SQ1 (though the direction is different), the rotor movement amount electrical angle θ _e increases in the positive direction, and at the same time, the variable gain K _th decreases by 1 every 10 [ms] from the initial value K _0. Also, as described above in (2), the control phase angle θ _ec changes by +90 deg from the end of sequence SQ1 and then decreases, starts to increase when the variable gain K _th becomes 1, and then maintains a substantially constant positive value. Also, when the rotor movement amount electrical angle θ _e becomes |C ₁ | or less, it is determined to be in a "zero movement amount" state for the sake of convenience, and the motor stop counter starts counting from that point.

In this sequence SQ2, if the rotor movement amount electrical angle _θe becomes equal to or greater than the set value _C2 , it is determined that the motor 50 is operating. If the rotor movement amount electrical angle _θe does not reach the set value _C2 , it is determined that the motor 50 is locked and cannot be retracted, and an alarm is generated and the retraction operation is stopped, thereby preventing erroneous estimation of the magnetic pole position.
Also, similarly to sequence SQ1, when the time corresponding to the count value of the motor stop counter reaches the set time _TS , it is determined that the retraction operation is completed (Yes in step S4 of FIG. 2).

In the sequence SQ2, for example, when the initial value K ₀ of the variable gain K _th is 1, the amount of movement θ _e2 of the control phase angle θ _ec after the above-mentioned set time T _S has elapsed is expressed as follows:
(1) When the amount of movement of the initial control phase angle of sequence SQ2 is θ _e1 +90 deg:
θ _e2 = -θ _a2 = -45 deg
(2) When the amount of movement of the initial control phase angle of sequence SQ2 is θ _e1 −90 deg:
θ _e2 = -θ _a2 = +45 deg
It becomes.
The phasor diagram on the right side of FIG. 4 shows the state at the end of sequence SQ2.

Returning to FIG. 2, after the set time T _S has elapsed in step S4, it is determined whether the accumulated amount of movement of the rotor is zero or not. If it is zero, it is determined that an abnormality has occurred in the magnetic flux position estimation operation, an alarm is output, and the estimation operation is forcibly terminated (steps S5 Yes, S5a in FIG. 2).
Moreover, if the accumulated amount of movement of the rotor is not zero (No in step S5), the process proceeds to the next sequence SQ3.

At the end of the above-mentioned sequence SQ2, in the actual motor, the estimated axis and the real axis may not completely match, and a slight phase difference may remain. This is because the current in the q-axis direction decreases as the phase difference, i.e., the magnetic pole position error, converges to near zero, and when the phase difference is near zero, the pull-in torque, which is proportional to this phase difference, is buried in the static friction of the motor.
Therefore, in order to make the phase difference zero, a final pull-in operation is performed in sequence SQ3 (step S6 in FIG. 2).

The final pull-in operation of this sequence SQ3 is an operation for pulling in the magnetic pole by increasing the d-axis current, and the control phase angle _θec is fixed.
6, the d-axis current command i _d ^* is set to a normal set value +100% (i.e., twice the normal set value) and the energization time is set to, for example, T _S /4. After the time T _S /4 has elapsed, the d-axis current command i _d ^* is set to zero and the process moves to generation of the electrical angle phase offset value Δθ in the next sequence SQ4 (step S7 in FIG. 2).

FIG. 5 is a phasor diagram at the end of sequence SQ3.
From FIG. 5, the control phase angle θ _est and the real axis phase angle θ _ang at the end of the sequence SQ3 are as follows:
θ _est = θ _e1 + 90 + θ _e2 ,
_θang = _θa1 + _θa2
Therefore, in sequence SQ4, the electrical angle phase offset value Δθ is calculated by the following formula.
θ _offset = θ _est - θ _ang = θ _e1 + 90 + θ _e2 - (θ _a1 + θ _a2 ) = Δθ

Since the electrical angle phase offset value Δθ corresponds to the magnetic pole position error, when the motor 50 is operated normally, each of the coordinate conversion units 31, ₃₂ performs coordinate conversion for current control and voltage control using a control phase angle θ _ec ' (= θ _ang + Δθ) obtained by adding the electrical angle phase offset value Δθ to the real axis phase angle θ ang, and the motor 50 can be controlled by a well-known method.
This makes it possible to control the motor 50 with high precision while eliminating magnetic pole position errors.

10: Pull-in current command generating unit 20: Current control unit 31: dq/uvw coordinate conversion unit 32: uvw/dq coordinate conversion unit 40: PWM calculation unit 50: AC synchronous motor 60: Pulse generator (PG)
70: Electrical angle calculation unit 80: Current detection unit

Claims (5)

1. A magnetic pole position estimation device that estimates an actual magnetic pole position of an AC synchronous motor based on a movement electrical angle obtained by converting a movement amount of a rotor or a mover of the AC synchronous motor when a predetermined current is caused to flow through an armature winding of the AC synchronous motor based on a current command,
a control phase angle obtained by multiplying the movement amount electrical angle by a predetermined variable gain, and using the control phase angle, a coordinate transformation is performed for feeding back a current flowing through the armature winding , and an electrical angle phase offset value to be added to the movement amount during normal operation of the AC synchronous motor is calculated. 2. The magnetic pole position estimation device for an AC synchronous motor according to claim 1,
4. The magnetic pole position estimation device for an AC synchronous motor, wherein the variable gain is a gain that decreases over time from a preset initial value to a lower limit value of 1. 3. The magnetic pole position estimation device for an AC synchronous motor according to claim 1,
a time-out period elapses after the predetermined current is passed, the time-out period elapses, and an alarm is generated to stop the magnetic pole position estimation operation. The magnetic pole position estimation device for an AC synchronous motor according to any one of claims 1 to 3,
a magnetic pole position estimating device for estimating a magnetic pole position of an AC synchronous motor, the magnetic pole position estimating device generating an alarm and stopping an operation of estimating the magnetic pole position when the amount of movement does not exceed a predetermined value. A control device for an AC synchronous motor, characterized in that the electrical angle phase offset value is calculated as a magnetic pole position error by the magnetic pole position estimation device described in any one of claims 1 to 4, and during normal operation of the AC synchronous motor, coordinate conversion for current control and voltage control is performed using a control phase angle obtained by adding the electrical angle phase offset value to the movement amount.

Images