JPH11299277A - Motor torque correction device and motor driving device provided with the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driving device provided with a torque correction device reducing the effects of torque ripples which a motor generates. SOLUTION: Storage means 3 and 4 storing and outputting two types of torque ripple data Tcg(θn) and Trp(θn), corresponding to a motor angle (0 degree <= θn<360 degrees) for one rotation are installed in a motor 1. Storage means 6 and 7 inputting and storing two types of torque ripple data outputted from the motor, a torque correction device 8, obtaining a torque correction signal Tm by Tm=G1.Tcg(θp)+G2.Trp(θp).Tref (where G1 and G2 are gains, θp is the prediction value of a motor angle, when the torque correction signal Tm is actually reflected on torque and Tref is a torque command) from an angle signal inputted from an angle detector 2 and torque ripple data and a means 9 subtracting the torque correction signal Tm from a torque command Tref are installed in a driving device 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motor torque correction device for correcting a torque ripple generated depending on an angle of a motor, a motor drive device including the device, and the like.

[0002]

2. Description of the Related Art Conventionally, torque ripples generated by a motor depending on the relative angle between a stator and a rotor include:
There are a cogging torque ripple, which is a fixed value regardless of the magnitude of the generated torque of the motor, and a type which is proportional to the generated torque. Such torque ripple causes uneven speed and position deviation of the servomotor, causes uneven feed on the feed shaft in the NC device, and lowers the cutting surface accuracy. In the positioning control, the response during settling changes. And the settling time is not constant. Conventional servo control has attempted to reduce the effect of ripple by increasing the loop gain of the speed loop and position loop in response to disturbance torque.However, excessively increasing the gain may cause mechanical resonance. There is a limit to the reduction effect, so recently,
For example, various schemes have been proposed to reduce the torque ripple without increasing the loop gain, as proposed in Japanese Patent Application Laid-Open No. 7-284286.

FIG. 4 is a block diagram of a conventional motor torque correction apparatus using a torque ripple correction method.
A control device 102 such as a device is a shared memory for receiving commands to the motor output from the NC device 101 and transferring the commands to the CPU of the digital servo circuit. Reference numeral 103 denotes a digital servo circuit, which includes a CPU, a ROM, a RAM, and the like, performs control of motor position, speed, current, and correction of torque ripple, corrects a torque command, and outputs a corrected torque. 104 is a servo amplifier constituted by an inverter or the like, 105 is a servomotor, 106 is a digital servo circuit 103 which detects the rotational position and speed of the motor 105.
This is a position / speed detector that feeds back the position and speed. Here, the torque ripple T1 is expressed as T1 (I, θ) from the current I and the angle θ of the motor, and the torque value T0, T0 = Kt · Tc + T1 (θ, I) generated by the motor before ripple correction is calculated by digital servo. The circuit 103 corrects the torque command Tc to correct the torque ripple. The torque command Tc is corrected to become a corrected torque command Tc '.
The torque value T0 after performing the correction in this manner becomes T0 = Kt · Tc ′ + T1 (θ, I), and the correction torque command Tc ′ is expressed as Tc ′ = Tc · (1 + A) where A is the correction term. Can be. The digital servo circuit 103 corrects and reduces the torque ripple T1 (θ, I).
A is formed so that A = −T1 (θ, I) / Kt · Tc, and Tc ′ is output in which the torque command Tc is corrected.

[0004]

However, in the above-mentioned conventional example, the torque command is corrected in accordance with the change in the torque ripple instead of increasing the speed loop gain. It is possible to correct the torque ripple, but basically, based on the detection data from the position / speed detector 106, etc.
Since the magnitude of T1 is extracted and the torque correction Tc 'is output to it, there is a time lag before the current torque correction Tc' is reflected in the actual torque, and the cycle changes periodically. There is a problem that accurate tracking and correction are not performed on the torque ripple. Therefore, the present invention decomposes and controls the torque ripple into a fixed value type torque ripple independent of the generated torque of the motor and a torque ripple component proportional to the generated torque, and the output torque correction signal is reflected in the actual torque. By performing predictive control to correct the torque ripple by predicting the motor angle at the time,
It is an object of the present invention to provide a motor torque correction device that enables highly accurate extraction and correction of torque ripple.

[0005]

According to a first aspect of the present invention, there is provided a torque ripple measuring device having a storage device, wherein a torque measurement signal applied to a motor shaft measured by a torque sensor and an angle measurement signal are output. From the motor angle signal detected by the detector, the motor angle for one rotation (0
N ≦ torque ripple data corresponding to degrees ≦ θn <360 degrees, n = 1, 2,..., N) is obtained and stored in the storage device. According to a second aspect of the present invention, there is provided a motor having an angle detector, further comprising storage means for storing and outputting torque ripple data which is an output of the torque ripple measuring device according to the first aspect. According to a third aspect of the present invention, in the motor according to the second aspect, the storage means uses a memory in the angle detector. Further, according to the present invention, the motor angle for one rotation (0 degree ≦ θn <360) obtained from the measurement signal of the torque applied to the motor shaft and the angle signal output from the motor angle detector.
It is characterized in that a torque correction signal is obtained from N pieces of torque ripple data corresponding to degrees (n = 1 2,..., N) and an angle signal from an angle detector. According to a fifth aspect of the present invention, in the motor torque correction device of the fourth aspect, the torque ripple data is obtained from the storage device of the motor of the second aspect. And
According to a sixth aspect of the present invention, in the motor torque correction device according to the fourth or fifth aspect, the torque correction signal Tm is obtained by the following equation. That is, Tm = G · Tcg
(Θp), where G is the gain and θp is the torque correction signal Tm
Is the predicted value of the motor angle at the time when it is actually reflected in the torque. According to a seventh aspect of the present invention, in the motor torque correction device according to the sixth aspect, the predicted value θp of the motor angle is obtained by the following equation. That is, θp = θ (t) + (dθ / dt) · dtm, where θ (t) is the motor angle at the detection time t, dθ / dt is the angular velocity, dtm
Is the time from the detection time t to the time when the torque correction signal reflects on the actual torque. According to an eighth aspect of the present invention, in the motor torque correction device according to the fourth or fifth aspect, the torque correction signal Tm is obtained by the following equation. That is, Tm = G · [w1 Tcg (θk) + w2 Tcg (θk +
1)] / (w1 + w2), where G is a gain, and θk and θk + 1 are θk <θ in the motor angle θn of the torque ripple data.
The motor angles satisfying p <θk + 1, w1 and w2 are interpolation coefficients.
Further, according to a ninth aspect of the present invention, in the motor torque correction device according to the fourth or fifth aspect, the torque ripple data is a fixed value type torque ripple data Tcg (θn) and a proportional torque ripple data Trp (
θn), and the torque correction signal Tm is obtained by the following equation from the two types of torque ripple data and the angle signal from the angle detector. That is, Tm = G1
Tcg (θp) + G2Trp (θp) Tref, where G1 and G2 are gains, θp is the predicted value of the motor angle at the time when the torque correction signal Tm is actually reflected in the torque, and Tref is the torque command . According to a tenth aspect of the present invention, there is provided a motor driving device for giving a torque command to a motor, wherein a storage means for inputting and storing torque ripple data, and a motor torque correction device according to any one of the fourth to ninth aspects. Means for reducing the output of the motor torque correction device from the torque command,
It is characterized by having. The invention according to claim 11 is the motor drive device according to claim 10, wherein
The storage means stores fixed-value type torque ripple data Tcg.
(Θn) and storage means for proportional torque ripple data Trp (θn) proportional to the generated torque. According to a twelfth aspect of the present invention, in the motor driving device according to the tenth or eleventh aspect, the input of the torque ripple data to the motor driving device is input from the storage means of the motor according to the second or third aspect. Features. According to a thirteenth aspect of the present invention, in the motor driving device according to the twelfth aspect, the input of the torque ripple data to the motor driving device is performed using a means for transmitting an angle signal. According to a fourteenth aspect of the present invention, in the motor driving device according to the twelfth or thirteenth aspect, input of torque ripple data to the motor driving device is performed by connecting the motor and the motor driving device and first turning on a power supply. It is characterized in that it is executed only the first time when it is thrown. Claim 15
Each of the inventions described below also has features, but details will be described in the following “Detailed description of the invention”.

As described above, according to the above configuration, the prediction control using the two types of torque ripple data Tcg and Trp stored in advance in the encoder (angle detector) enables the
The system performs high-precision torque ripple correction control decomposed into systems, and when generating and outputting a torque correction signal, predicts and calculates the predicted value θp of the motor angle at the time that is actually reflected in the torque, and calculates the time lag of the torque ripple correction. Accurate correction control is possible without it. Specifically, the torque correction signal Tm is calculated as follows: Tm = G1 · Tcg (θp) + G2 · Trp (θp) ·
Tref, where G1 and G2 are gain, θp is torque correction signal
The predicted value of the motor angle at the time when Tm actually reflects on the torque, and Tref is calculated and formed as a torque command. The first term in the above equation is a correction term for the torque ripple Tcg independent of the generated torque, and the second term is a correction term for the torque ripple Trp proportional to the generated torque. The predicted value of the motor angle θp is represented by θ (t) for the motor angle at the motor angle detection time t, dθ / dt for the angular velocity, and dtm for the detection time t.
The time until the time when the torque correction signal Tm reflects on the actual torque can be generally predicted by θp = θ (t) + dθ / dt · dtm. However, when the sampling cycle is Ts, the sampling time i・ When the motor angle θ (iK) of K (≧ 0) sampling past in Ts is input from the encoder, the time is m sampling ahead of time i · Ts.
Using the predicted value of the motor angle at (i + m) · Ts, the predicted value θp is calculated as θp = θ ^* (i + M) using the predicted value of the motor angle θ ^* (i + m). Becomes possible. Alternatively, a measuring device that measures torque ripple and outputs it to the motor side, a motor equipped with a memory for storing torque ripple data in an encoder, and reads torque ripple data stored from the motor side,
By combining with a drive device that generates a torque correction signal in consideration of the time delay of the torque ripple correction, it is possible to collect torque ripple data for each individual motor,
With a low-cost configuration that requires only one memory for storing torque ripple data, a torque correction device capable of predicting the motor angle and eliminating a time lag of the torque ripple correction signal to enable accurate correction can be configured.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram of a motor torque correction device according to a first embodiment of the present invention. FIG.
, 1 is a motor, to which an angle detector 2 is attached. The angle detector 2 is for detecting the angle θ from the origin angle of the motor, and may be directly detected by an absolute encoder or indirectly detected by counting from the origin pulse using a relative encoder. Is also good.
3 and 4 are generated by the motor 1 depending on the motor angle 2
This is a memory in which various types of torque ripple data are stored, and a motor angle for one rotation (0 degree ≦ θn <36
0 degrees, n = 1, 2,..., N) and N torque ripple data Tcg (θn), n = 1, 2,.
(θn), n = 1, 2,..., N are stored. Here, the ripple data Tcg which does not depend on the torque generated by the motor such as the cogging torque is stored in the memory 3, and the ripple data Trp proportional to the generated torque is stored in the memory 4. A transmitter 10 transmits the angle signal θ, and also functions as a transmitter for two types of torque ripple data. Reference numeral 5 denotes a driving device for driving the motor 1. The driving device 5 inputs the two types of torque ripple data stored in the memories 3 and 4 via the transmitter 10 and respectively inputs
And 7 are stored. Reference numeral 8 denotes a torque correction device,
And 7, a torque correction signal Tm is calculated and output from the torque ripple data stored in the angle detector 2, the angle signal θ input from the angle detector 2 via the transmitter 10, and the torque command Tref. Reference numeral 9 denotes a subtractor for subtracting the torque correction signal Tm from the torque command Tref, and the motor 1 is driven using the output Tref 'as a corrected torque command. As a result, the torque command is subtracted by the amount of the torque ripple generated by the motor 1, so that the torque ripple can be significantly reduced.

Next, the operation will be described focusing on the prediction calculation with such a configuration. First, the torque correction signal Tm
Is the torque ripple data Tc stored in the memories 6 and 7.
g (θn), n = 1,2, ..., N and Trp (θn), n = 1,2, ...,
It is given by the following equation (1) using N. Tm = G1 · Tcg (θp) + G2 · Trp (θp) · Tref (1) where G1 and G2 are gains, and θp is the torque correction signal Tm
Is a predicted value of the motor angle at the time when the torque is reflected on the actual torque, and Tref is a torque command. The predicted value θp can be obtained by the following equation. θp = θ (t) + (dθ / dt) · dtm (2) where θ (t) and dθ / dt are the motor angle and the angular velocity at the detection time t, and are transmitted from the angle detector 2 through the transmitter 10. It can be obtained from the input angle signal. dtm is the time from the detection time t to the time at which the torque correction signal Tm reflects on the actual torque. When the sampling period is Ts in the driving device 5 and the motor angle θ (iK) of K (≧ 0) sampling past at the sampling time i · Ts is input from the angle detector 2 via the transmitter 10, M sampling time from time i · Ts (i + m) · T
using the predicted value of the motor angle θ ^* (i + m) in s, the predicted value θp is calculated by the following equation θp = θ ^* (i + M), or θp = θ ^* (i + M-1) (3) Alternatively, interpolation can be given by θp = θs1 ・ θ ^* (i + M−1) + s2 ・ θ ^* (i + M)｝ / (s1 + s2) (4). Where M is M-1 ≤ dtm '/ Ts ≤
M is an integer, dtm 'is the time from the sampling time i · Ts to the time when the torque correction signal reflects on the actual torque, and s1 and s2 are interpolation coefficients. In particular, when K = 0 and M = 1, it is given by θp = ｛s1) θ (i) + s2iθ ^* (i + 1)｝ / (s1 + s2) (5). Here, the motor angle predicted value θ ^* (i + m) at m sampling destinations is given by the following equation (6) or (7),

[0009]

(Equation 9)

[0010]

(Equation 10)

Alternatively, the following equation

[0012]

[Equation 11]

And one of the following four equations.

[0014]

(Equation 12)

Here, Δ represents an increment value during the sampling period Ts.

Hereinafter, equations (6) to (12) will be described. First, equations (6) and (7) are derived. Now input any of the torque command, speed command or position command of the motor
and the transfer function model from this input u (i) to the motor angle θ (i) is Gry (z) = (b ₁ z ⁻¹ +... + b _Nb
z ^-Nb ) / (1-a ₁ z ^-1 − ・ ・ ・ −a _Na z ^-Na ) ^Assuming that it is obtained in the discrete-time system by the z-transform of (13), the input-output model is Become.

[0017]

(Equation 13)

At time i · Ts (hereinafter referred to as time i for convenience), the model estimation value θ ′ of the motor angle after time iK
When (i + m) (m ≧ −K + 1) is represented using the actually measured value θ (in) (n ≧ K), the following expression is obtained.

[0019]

[Equation 14]

Here, the coefficients a'mn and b'mn are given by the following equations.

[0021]

(Equation 15)

Where an = 0 (n> Na), bn = 0 (n> Nb), b'mn =
0 (n <1) Then, the motor angle after time iK is

[0023]

(Equation 16)

Predicting the above gives the above equation (6). Here, the coefficients Amn and Bmn are given by the following equation and equation (16). AmK = 1 + a'mK n = K Amn = a'mn-a (nK) K + 1 ≦ n ≦ Na + K Bmn = b'mn-b (nKm) 1 ≦ n ≦ Nb + K + m (18 ) Where bn = 0 (n <1), a'm (Na + K) = 0, b'm (Nb + K + m) = 0 and future input is u (j) = u (i) ( j> i), the same equation (7) is obtained, and the coefficient Amn is expressed by the equations (16) and (18).
Is given by the following equation.

[0025]

[Equation 17]

Where an = 0 (n> Na), bn = 0 (n <1 and n>
Nb), b'm (Nb + K) = 0 Next, equations (9) and (10) are derived. Now input increment value Δu
Discrete-time transfer function model from (i) to motor angle increment Δθ (i) Gdd (Z) = (b ₁ z ⁻¹ +... + b _Nb z ^−Nb ) / (1-a ₁ z ⁻¹⁾ − ・ ・ ・ −a _Na z ^-Na ) (20), and at time i, the motor angle increment after time iK is calculated as

[0027]

(Equation 18)

The above equation (9) is obtained by the prediction. here,
The coefficients Asn and Bsn are given by the following equations.

[0029]

[Equation 19]

Where an = O (n> Na), bn = 0 (n> Nb), Bsn = 0
(n <1) Also, if the future input is similarly predicted as u (j) = u (i) (j> i), the above equation (10) is obtained.The coefficient Asn is expressed by the equation (22), and Bsn is Given by the formula.

[0031]

(Equation 20)

Where an = O (n> Na), bn = 0 (n <1 and n>
Nb) Further, the equations (11) and (12) are derived. Now, a discrete-time transfer function model from the input u (i) to the motor angle increment Δθ (i) Grd (Z) = (b ₁ z ^-1 +... + B _Nb z ^-Nb ) / (1-a ₁ z ^-1 − ・ ・ ・ −a _Na z ^-Na ) (24), and at time i, the motor angle increment value after time iK is calculated as

[0033]

(Equation 21)

The above equation (11) is obtained by the prediction. here,
The coefficients Asn and Bsn are given by the following equations.

[0035]

(Equation 22)

Where an = O (n> Na), bn = 0 (n> Nb), Bsn = 0
(n <1) Also, if the future input is similarly predicted as u (j) = u (i) (j> i), the above equation (12) is obtained.The coefficient Asn is expressed by the equation (26), and Bsn is Given by the formula.

[0037]

(Equation 23)

Where an = 0 (n> Na) and bn = 0 (n <1 and n> N
b) Also, when the predicted value θp obtained by any of the above methods does not match the angle θn, n = 1, 2,..., N in the torque ripple data, that is, n = k, k + 1 Angle θk, θk + 1
When θk <θp <θk + 1, the value closer to θp is used. For example, if it is θk, the torque correction signal
Tm may be Tm = G1Tcg (θk) + G2Trp (θk) Tref (28), or Tm = G1 ｛w1 Tcg (θk) + w2 Tcg (θk + 1)｝ / (w1 + w2) + G2 ・ ｛w1 Trp (θk) + w2 Trp (θk + 1)｝ ・ Tref / (w1 + w2) (29) w1 and w2 are interpolation coefficients. Good. As described above, according to the first embodiment, when the torque ripple is decomposed into Tcg and Trp, and the reduction control is performed, the motor angle predicted value θ ^* (i + m) is obtained by a high-level calculation, so that it is accurate. A simple torque ripple correction is performed.

Next, a second embodiment of the present invention will be described with reference to the drawings. FIG. 2 is a configuration diagram of a motor torque ripple measuring device used in the present invention. FIG. 3 is a configuration diagram of a motor torque correction device according to a second embodiment of the present invention. In FIG. 2, reference numeral 11 denotes a motor to which an angle detector 12 is attached. The angle detector 12 has a built-in memory 14 for detecting the angle θ from the origin angle of the motor, and may be directly detected by an absolute encoder or indirect by counting from the origin pulse using a relative encoder. May be detected. Reference numeral 13 denotes a torque ripple measuring device, which inputs a measurement signal of a torque applied to a motor shaft and an angle signal θ output from the angle detector 2 and receives a motor angle for one rotation (0 degrees ≦ θn <360 degrees;
N = 1, 2,..., N) corresponding to N torque ripple data Tcg (θn), n = 1, 2,. Many methods can be considered for measuring the torque applied to the motor shaft. In this example, the motor 11 is rotated at a constant low speed by the constant-speed rotating device 18 via the connecting portion 19.
The torque applied to the connecting portion or the motor shaft is measured by a torque sensor (such as an electromagnetic type) 17. A transmitter 16 transmits the angle signal θ, and also serves as a transmitter for torque ripple data for one rotation. The torque ripple data for one rotation stored in the memory 15 is transmitted to the angle detector 12 via the transmitter 16 and stored in the memory 14.

FIG. 3 shows the simplified motor torque correction device of FIG. While the motor torque compensator of FIG. 1 handles two types of torque ripple data, the one of FIG. 3 handles only fixed value type torque ripple data Tcg (θn) among the two types of torque ripple data. I have. This is Figure 1
Compared with the motor torque correction device described above, there is an advantage that the speed of the correction is faster and the memory can be saved, so that the cost is reduced, and the accuracy of the motor torque correction is slightly reduced, but there is no practical problem. In particular, a sufficient reduction effect can be obtained for torque ripple correction at low speed. In FIG. 3, reference numeral 21 denotes a driving device for driving the motor 11. The driving device 21
When power is turned on, torque ripple data for one rotation stored in the memory 14 of the angle detector 12 is input via the transmitter 16 and stored in the memory 22. A torque correction device 23 calculates and outputs a torque correction signal Tm based on the torque ripple data stored in the memory 22 and the angle signal θ input from the angle detector 12 via the transmitter 16. 24 is a subtractor for subtracting the torque correction signal Tm from the torque command Tref. The motor is driven using the output Tref 'as a corrected torque command. As a result, the torque command is subtracted by the amount of the torque ripple generated by the motor, so that the influence of the torque ripple is greatly reduced. The transmission of the torque ripple data from the memory 14 to the memory 22 may be performed every time the power is turned on. However, the transmission is performed only once when the motor and the driving device are connected and the power is first turned on. This may not be performed unless the motor or the driving device is changed.

Next, the operation will be described focusing on the method of calculating the torque correction signal. The torque correction signal Tm is stored in the memory 2
2, the torque ripple data Tcg (θn), n = 1, 2,
.., N are given by the following equation. Tm = G · Tcg (θp) (30) Here, G is a gain, and θp is a predicted value of the motor angle at the time when the torque correction signal Tm is reflected on the actual torque. The predicted value θp is obtained by the following equation. θp = θ (t) + (dθ / dt) · dtm (31) where θ (t) and (dθ / dt) are the motor angle and the angular velocity at the detection time t. It can be obtained from the angle signal input through the camera. dtm
Is the time from the detection time t to the time at which the torque correction signal Tm reflects on the actual torque. If the predicted value θp does not match the torque ripple measurement angle θn, n = 1, 2,..., N, that is, θk <θp <θk + 1 for the torque ripple measurement angles θk, θk + 1 In this case, the value closer to θp is used, for example, if it is θk, the torque correction signal Tm
Can be set as Tm = G ・ Tcg (θk) (32), or Tm = G ・ ｛w1 Tcg (θk) + w2 Tcg (θk + 1)｝ / (w1 + w2) (33) You may ask. As described above, according to the second embodiment, the torque ripple of each motor can be measured and detected by actual measurement regardless of the manufacturer's specifications and the like. For example, the torque ripple corresponding to the motor angle can be determined based on the obtained torque ripple data. A table of data, its rate of change, a corresponding torque correction signal and the like is created and placed, and the motor angle at the time when the torque correction signal is reflected is predicted, and the torque correction signal at the predicted time is read from the table and corrected. Such a correction method is possible, so that the worst case in which the torque ripple is increased in the direction of increasing the torque ripple due to the time delay of the correction signal does not occur, and the timing control that can accurately reduce and reduce the torque ripple can be performed. Also, as in the case of the first embodiment,
Tcg and Trp are not decomposed and suppressed, but actual measurement and predictive control of ripple can provide a sufficient reduction effect especially for torque ripple correction at low speeds, and saves memory, resulting in a low-cost correction system. it can. In the first embodiment, two types of torque ripple data Tcg, Trp are stored in the memory in advance, and the process is started. However, the measurement is started by providing a ripple measuring device as in the second embodiment. Of course, it is possible.

[0042]

As described above, according to the present invention, the torque ripple is decomposed into two types of torque ripples, Tcg and Trp, and high-precision correction control is performed. Is greatly reduced, and speed control or position tracking control that does not cause speed ripple and tracking error can be performed.
Furthermore, since the torque ripple data of each motor can be sampled by actual measurement and the torque ripple is accurately suppressed by the predictive control, the speed control and the position which do not cause the speed ripple or the tracking error especially at a low speed are realized. There is an effect that tracking control can be performed.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a motor torque correction device according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of an apparatus for measuring torque ripple of a motor according to the present invention.

FIG. 3 is a configuration diagram of a motor torque correction device according to a second embodiment of the present invention.

FIG. 4 is a configuration diagram of a conventional motor torque correction device.

[Explanation of symbols]

 1,11 Motor 2,12 Angle detector 3,4,6,7,14,15,22 Memory 5,21 Motor drive 8,23 Torque corrector 9,24 Subtractor 10,16 Signal transmitter 13 Ripple measurement Device 17 Torque sensor 18 Constant-speed rotating device 19 Connecting part

Claims (21)

[Claims] 1. A torque ripple measuring device having a storage device, comprising: a motor angle for one rotation (a motor angle signal) based on a torque measurement signal on a motor shaft measured by a torque sensor and a motor angle signal detected by an angle detector. 0 degrees ≦ θn <360 degrees, n = 1,
(2,..., N), and obtains the torque ripple data and stores it in the storage device. 2. A motor provided with an angle detector, comprising a storage means for storing and outputting torque ripple data which is an output of the torque ripple measuring device according to claim 1. 3. The motor according to claim 2, wherein said storage means uses a memory in said angle detector. 4. A motor angle for one rotation obtained from a measurement signal of a torque applied to a motor shaft and an angle signal output from a motor angle detector (0 degrees ≦ θn <360 degrees, n = 1 2,
.. N) corresponding to the torque ripple data,
A motor torque correction device, wherein a torque correction signal is obtained from an angle signal from an angle detector. 5. The motor torque correction device according to claim 4, wherein the torque ripple data is obtained from the storage device for the motor according to claim 2. 6. The motor torque correction device according to claim 4, wherein the torque correction signal Tm is obtained by the following equation. Tm = G · Tcg (θp) where G is the gain, θp is the predicted value of the motor angle at the time when the torque correction signal Tm is actually reflected in the torque, 7. The motor torque correction device according to claim 6, wherein the predicted value θp of the motor angle is obtained by the following equation. θp = θ (t) + (dθ / dt) · dtm where θ (t) is the motor angle at the detection time t, dθ / dt is the angular velocity, and dtm is the torque correction signal reflected on the actual torque from the detection time t. Time until the time. 8. The motor torque correction device according to claim 4, wherein the torque correction signal Tm is obtained by the following equation. Tm = G · [w1 Tcg (θk) + w2 Tcg (θk + 1)] / (w1 + w2) where G is a gain, and θk and θk + 1 are motor angles θ of the torque ripple data.
In n, motor angles satisfying θk <θp <θk + 1, w1 and w2 are interpolation coefficients. 9. The motor torque correction device according to claim 4, wherein the torque ripple data is fixed value type torque ripple data Tcg (θn) and proportional torque ripple data Trp (θn) proportional to generated torque. A motor torque correction device, wherein a torque correction signal Tm is obtained from the following two types of torque ripple data and an angle signal from an angle detector by the following equation. Tm = G1Tcg (θp) + G2Trp (θp) Tref where G1 and G2 are gains, θp is the predicted value of the motor angle at the time when the torque correction signal Tm is actually reflected in the torque, and Tref is the torque. Command. 10. A motor drive device for giving a torque command to a motor, a storage means for inputting and storing torque ripple data, a motor torque correction device according to any one of claims 4 to 9, and a motor torque correction device. Means for reducing the output of the device from the torque command. 11. The motor drive device according to claim 10, wherein said storage means stores fixed-value type torque ripple data Tcg.
A motor drive device comprising: storage means for (θn); and storage means for proportional torque ripple data Trp (θn) proportional to the generated torque. 12. The motor driving device according to claim 10, wherein the input of torque ripple data to the motor driving device is input from the storage means of the motor according to claim 2 or 3. apparatus. 13. The motor driving device according to claim 12, wherein the input of the torque ripple data to the motor driving device is performed using a means for transmitting an angle signal. 14. The motor drive device according to claim 12, wherein the input of torque ripple data to the motor drive device is performed at the first time when the motor and the motor drive device are connected and power is first turned on. A motor drive device characterized by performing only the following. 15. The motor torque correction device according to claim 6, wherein the sampling period is Ts and the sampling time i · Ts
K (≧ 0) sampling at past motor angle θ (ik)
A motor torque correction device characterized in that the predicted value θp of the motor angle is obtained by the following equation using the following equation: θp = θ ^* (i + M) or θp = θ ^* (i + M-1) where θ ^* (i + m) is a torque command, speed command, or position command, and Sampling time i · Ts predicted using a discrete-time transfer function model from the input u or the increment value Δu to the motor angle θ or the increment value Δθ
The predicted value of the motor angle at time (i + m) · Ts, which is m sampling ahead, Δ is the increment between sampling periods Ts, M is an integer that satisfies M-1 ≦ dtm '/ Ts ≦ M, dtm' is Time from the sampling time i · Ts until the torque correction signal is reflected in the actual torque. 16. The motor torque correction device according to claim 6, wherein the predicted value θp of the motor angle is obtained by the following equation. θp = ｛s1 ・ θ ^* (i + M-1) + s2 ・ θ ^* (i + M)｝ / (s1 + s2) where s1 and s2 are interpolation coefficients. θ ^* (i + m) considers any of the torque command, the speed command, and the position command as an input u, and uses a discrete-time transfer function model from the input u or the increment value Δu to the motor angle θ or the increment value Δθ. M more than sampling time iTs
Predicted value of motor angle at sampling destination time (i + m) Ts, Δ is increment value between sampling periods Ts, M is an integer that satisfies M-1 ≦ dtm '/ Ts ≦ M, dtm' is sampling time i・ Time until the torque correction signal is reflected on the actual torque from Ts. 17. The motor torque correction device according to claim 16, wherein, particularly when k = 0 and M = 1, the predicted value θp of the motor angle is obtained by the following equation. . θp = ｛s1 ・ θ (i) + s2 ・ θ ^* (i + 1)｝ / (s1 + s2) 18. The motor according to claim 15, wherein the motor angle predicted value θ ^* (i + m) at the m sampling destinations is obtained by the following equation. Torque correction device. (Equation 1) Or, Here, Amn and Bmn are coefficients determined by a discrete-time transfer function model from the input u to the motor angle θ. 19. The motor torque correction device according to claim 15, wherein the motor angle prediction value θ ^* (i + m) at the m sampling destinations is obtained by the following equation. Torque correction device. (Equation 3) And Or, Here, Asn and Bsn are coefficients determined by a discrete time transfer function model from the input increment value Δu to the motor angle increment value Δθ. 20. The motor according to claim 15, wherein the motor angle prediction value θ ^* (i + m) at the m sampling destinations is obtained by the following equation. Torque correction device. (Equation 6) And Or, Here, Asn and Bsn are increment values Δ of the motor angle from the input u.
A coefficient determined by a discrete-time transfer function model up to θ. 21. The motor torque correction device according to claim 4, wherein the torque ripple data is fixed value type torque ripple data Tcg (θn) and proportional torque ripple data Trp (θn) proportional to the generated torque. A motor torque correction device, wherein a torque correction signal Tm is obtained from the following two types of torque ripple data and an angle signal from an angle detector by the following equation. Tm = G1 ・ ｛w1 Tcg (θk) + w2 Tcg (θk + 1)｝ / (w1 + w2) +
G2 ・ ｛w1 Trp (θk) + w2 Trp (θk + 1)｝ ・ Tref / (w1 + w
2) Here, G1 and G2 are gains, and θk and θk + 1 are motor angles θ of the torque ripple data.
In n, motor angles satisfying θk <θp <θk + 1, w1 and w2 are interpolation coefficients, and Tref is a torque command.