JP2003023785A - Servo motor control device and control method - Google Patents

Abstract

(57) [Problem] To provide a servomotor control device for performing position control, in which a position deviation of an object to be controlled having a large viscous friction is shifted near zero during operation, and after a movement command is zero. Another object of the present invention is to shorten the positioning settling time by setting the position deviation without largely changing the position deviation. A servo motor control device uses a position deviation integrator 7 and a speed deviation integrator 13 in combination, receives a signal from a timing generator 17 near a position where a movement command becomes "0", and receives a position deviation integrated value. By reducing the velocity deviation integral value, the occurrence of position deviation during acceleration / deceleration can be suppressed, and the position deviation can be maintained near zero even after the movement command is zero, so that the settling time can be shortened.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a servomotor control device and control method for position control, and more particularly to reduction of positioning settling time in a servomotor control device.

[0002]

2. Description of the Related Art As a conventional servo motor control device, a device disclosed in Japanese Patent Laid-Open No. 10-105247 is known. A conventional servo motor control device will be described with reference to FIGS. 24 and 25. FIG. 24 is a control block diagram of a conventional servo motor control device. FIG. 25 is FIG.
6 is a flowchart showing the operation of the control device shown in FIG. 4, showing position loop processing and speed loop processing.

As shown in FIG. 24, the position command generator 10
Position command θ * from 4 is servo driver 1 at regular intervals
It is configured to control the electric current to the servo motor 101 by being input to 03 and processed. The servo driver 103 supplies a drive current to the servo motor 101 so as to match the position command θ *, and performs position control.
The servo motor 101 is provided with a position detector 102 for detecting the rotation angle of the servo motor 101. The position signal θ output from the position detector 102 is input to the servo driver 103 at regular intervals. The servo driver 103 includes a position deviation calculator 105, a position deviation proportional calculator 106, a speed calculator 110, and a speed deviation calculator 11
1, speed deviation proportional unit 112, speed deviation integrator 113, current command calculator 114, current controller 115, and timing generator 117. In FIG. 24, Kpp is a position proportional gain, Kvp is a velocity proportional gain, K
vi is a velocity integral gain, and s is a Laplace operator.

The operation of the conventional servo motor control device configured as described above will be described with reference to FIGS. 24 and 25. The position detector 102 detects the rotation angle of the servo motor 101 at regular intervals and outputs the position signal θ to the servo driver 1
Output to 03. Further, the position command θ is input to the servo driver 103 from the position command generator 104 at regular intervals (process step 101 in FIG. 25). Next, the position loop processing (processing step 102) and the velocity loop processing (processing step 103 to processing step 109) are executed, and the servo driver 103 causes the position command generator 10 to operate.
Servo motor 101 so as to match the position command θ * of 4
A drive current is supplied to the position control device.

Next, the arithmetic processing in the servo driver 103 for controlling the drive of the servo motor as described above will be described. First, the position signal θ from the position detector 102 and the position command θ * from the position command generator 104 are input to the position deviation calculator 105 and subjected to subtraction processing. The position deviation calculator 105 calculates θe = θ * −θ for each position control cycle and outputs the position deviation θe. Then, the position deviation proportional device 6 calculates the product of the position deviation θe and the position proportional gain Kpp, and outputs the speed command ω *. In this way, the processing for calculating the speed command ω * is the position loop processing (processing step 102).
Is. Next, in processing step 103, the timing generator 117 executes the position command generator 1 as the first processing.
Based on the position command θ * output from 04, it is determined whether or not the movement command is “0”. If the movement command is not "0", the flag F is set to "0" in processing step 107. Here, the movement command is a command input to the position command generator 104, and is converted into a position command by counting the pulses input in the position command generator 104.

Next, the speed calculator 110 is the position detector 10
The position signal .theta. Output from 2 at a constant cycle is converted into a speed by the backward difference, and the speed signal .omega. Is output to the speed deviation calculator 111 by performing a filter process as necessary. Then, the speed deviation calculator 111 calculates ωe = ω * −ω for each speed control cycle using the speed command ω * and the speed signal ω, and calculates the speed deviation ωe as the speed deviation proportional unit 112 and the speed deviation integral. Output to the device 113. Speed deviation proportionalizer 112
Calculates and outputs a speed deviation proportional value which is a product of the speed deviation ωe and the speed proportional gain Kvp. In addition, the speed deviation integrator 1
13 is the integrated value Σωe of the speed deviation ωe and the speed integral gain Kvi
The speed deviation integral value Σωe (n), which is the product of and, is calculated and output (processing step 108). The calculation in the actual processing step 108 is expressed by the following equation (1), where Σωe (n) is the current speed deviation integrated value and Σωe (n-1) is the previous speed deviation integrated value.

[0007]

[Equation 1]

Then, the speed deviation proportional value calculated by the speed deviation proportionalizer 112 and the speed deviation integrated value calculated by the speed deviation integrator 113 are the current command calculator 11
4 is added and output as the current command value I * (processing step 109). In the current command calculator 114, the following equation (2)
The calculation indicated by is performed to calculate the current command value I *.

[0009]

[Equation 2]

In the above arithmetic processing, processing step 10
8 and the processing of processing step 109 correspond to the speed loop processing. Then, the current controller 115 controls the servo motor 101.
Current of the current command value I * is supplied to the servo motor 10
1 is controlled (processing step 110). Hereinafter, as long as a pulse is input from the position command generator 104 and there is a movement command, processing step 101 → processing step 102 → processing step 103 → processing step 107 → for each processing cycle.
The position loop processing and the speed loop processing of processing step 108-> processing step 109-> processing step 110 are repeatedly executed. In this way, the position loop processing and the speed loop processing are repeatedly executed, and the position and speed of the servo motor 101 are feedback-controlled. In the above process, the timing generator 117 determines the presence / absence of the movement command in the processing step 103. If the movement command disappears and the movement command is determined to be "0", the flag F is "1" in step 104. Determine whether or not. Since the flag F is initially set to “0” in processing step 107, processing steps 104 to 105 are executed.
Move to.

The position deviation θe read in the position loop processing in the processing step 102 is compared with a preset threshold value ε in the processing step 105. Position deviation θ
When the absolute value of e is larger than the threshold value width ε, the normal integration processing is performed in processing step 108 corresponding to the case where the timing signal is not output from the timing generator 117. Then, the current command value I * is calculated in processing step 109, and this current command value I * is sent to the current controller 115 in processing step 110. That is, when the absolute value of the position deviation θe is larger than the threshold width ε, normal speed loop processing is executed. Therefore, as long as the absolute value of the position deviation θe is larger than the threshold value ε, the processing step 101 → the processing step 102 → the processing step 103 → the processing step 10
The process of 4 → processing step 105 → processing step 108 → processing step 109 → processing step 110 is repeated.

On the other hand, the absolute value of the position deviation θe is the threshold value ε.
If it is equal to or smaller than the threshold value ε, the process proceeds to the processing step 106. The above processing is performed by outputting a timing signal from the timing generator 117 to the speed deviation integrator 113. That is, the speed deviation integrator 113
When the timing signal is input, the speed deviation integrated value Σωe (n-1) up to the previous cycle is multiplied by the setting coefficient k3 to obtain the speed deviation integrated value Σωe (n) for the cycle. Where coefficient k3
Is set to a value greater than 0 and less than 1,
The calculated speed deviation integrated value Σωe (n) has a small value.
The arithmetic expression of the processing step 106 is shown in the following expression (3).

[0013]

[Equation 3]

Then, after the flag F is set to "1" in the processing step 106, the processing step 109 and the processing step 110 are executed. As described above, since the flag F is set to "1" in the processing step 106, even if the movement command is "0" in the next control loop, the processing shifts from the processing step 104 to the processing step 108. Thus in the control loop,
The speed deviation integrated value Σωe (n) at processing step 106 only once
Is switched to a small value, the current command value I * becomes small, and the output torque of the servomotor 101 also becomes small. As a result, the occurrence of overshoot of the position deviation after the movement command becomes "0" is prevented. Furthermore, JP-A-4-233608 and JP-A-4-184512.
In the publication, as in the above-mentioned prior art, it is described that the accumulated velocity deviation integral value is reduced by switching the integral calculation of the velocity loop from the complete integral calculation to the incomplete integral calculation when the movement command is zero. Has been done. This allows
When the movement command is zero, that is, when the stop command is input,
The occurrence of overshoot is prevented.

[0015]

The structure of the conventional control device is a structure in which a position deviation occurs during operation. Therefore, it is an object of the conventional control device to eliminate the overshoot by reducing the velocity integral term when the position deviation occurs even when the movement command is zero. However, in order to shorten the settling time, a velocity feedforward term is inserted to converge the position deviation to near zero during operation, and when the movement command is zero, the position deviation is set as it is while maintaining the settling width near zero. In the case of developing into the system, there are the following problems only by the control structure and the reduction calculation processing of the velocity deviation integrated value shown in the above-mentioned prior art.

1. When viscous friction is large, a steady deviation occurs during operation, and a position deviation also occurs when the movement command becomes 0 (when the movement command is zero), so the settling time cannot be shortened. 2. If the velocity deviation integral term is suddenly reduced when the movement command is zero, the torque changes abruptly and vibration of the position deviation occurs, so the settling time cannot be shortened. 3. When the velocity feedforward term is used to shorten the settling time, the position deviation overshoots only by reducing the velocity deviation integral term due to the delay output of the velocity feedforward filter. In addition, overshoot may occur due to a speed detection filter or a delay in calculation. 4. In the case of a controlled object that moves vertically and gravity acts,
Even if the position deviation changes in the vicinity of zero, if the integral term that compensates for gravity when the movement command is zero is decreased, a position deviation occurs, so the settling time cannot be shortened. That is, in the conventional configuration, the position deviation cannot be close to zero during operation when controlling a control target having large viscous friction or the like, or when using a velocity detection filter or a velocity feedforward filter. Further, even if the position deviation changes near zero, if the value of the speed deviation integrator is simply reduced when controlling the vertical movement device, etc., the position deviation becomes large after the movement command becomes zero and the settling time becomes long. There was a problem.

The present invention solves the above-mentioned problems in the conventional control device, and for a control object having a large viscous friction or a control object on which a gravitational term acts, or for a control structure on which a filter acts. It is an object of the present invention to provide a servomotor control device and control method that can shorten the positioning settling time by causing the position shift to move near zero deviation during operation and settling without greatly changing the position deviation even after the movement command is zero. .

[0018]

In order to achieve the above object, a servomotor control device according to the present invention is configured to include a position command generator and a change amount of a position command output from the position command generator. A timing generator that generates a timing signal in the vicinity of a certain movement command becomes zero, and an integral calculation of the position deviation that is the difference between the movement command and the position of the servo motor is performed to output the position deviation integrated value. A position deviation integrator that decreases the position deviation integrated value when a timing signal from a generator is input, and an integral calculation of a speed deviation that is a difference between the speed command obtained from the movement command and the speed of the servo motor. To output the speed deviation integrated value and reduce the speed deviation integrated value when the timing signal from the timing generator is input. As a result, the servo motor control device of the present invention can significantly reduce the positioning settling time by performing settling without greatly changing the position deviation even after the movement command is zero.

According to another aspect of the present invention, there is provided a servo motor control device, which includes at least one position command generator and at least a displacement command output from the position command generator, which is a displacement command near zero. A timing generator that generates a timing signal and an integral calculation of the position deviation that is the difference between the movement command and the position of the servo motor are performed to output the position deviation integrated value, and the timing signal from the timing generator is input. The position deviation integrator that decreases the position deviation integrated value when operated, calculates the speed feedforward value based on the movement command, and decreases the speed feedforward value when the timing signal from the timing generator is input. And a velocity feedforward device for controlling the velocity. As a result, the servo motor control device of the present invention can shorten the settling time without overshooting the position deviation even after the movement command is zero.

In addition, according to another aspect of the invention, there is provided a servo motor control device, wherein a position command generator and a timing before a movement command, which is a variation of the position command output from the position command generator, become zero. A timing generator that generates a signal and an integral calculation of the position deviation that is the difference between the movement command and the position of the servo motor are performed to output the position deviation integrated value, and the timing generation is performed before the movement command becomes zero. A position deviation integrator that reduces the position deviation integrated value when a timing signal from the device is input, and a speed feedforward device that calculates a speed feedforward value based on the movement command and outputs a feedforward value. To do. As a result, the servo motor control device of the present invention,
Even after the movement command is zero, the settling time can be shortened without overshooting the position deviation.

Further, according to another aspect of the present invention, there is provided a servo motor control device, wherein a position command generator and a timing near a movement command, which is a change in the position command output from the position command generator, become zero. A timing generator that generates a signal, performs an integral calculation of a speed deviation that is the difference between the speed command obtained from the movement command and the speed of the servo motor, outputs a speed deviation integrated value, and outputs the speed deviation from the timing generator. Speed deviation integrator that decreases the speed deviation integrated value when the timing signal of (4) is input, and speed feedforward that calculates the speed feedforward value by performing the filtering process of the time constant preset based on the movement command. The speed of the controller and the servo motor using a filter time constant similar to the filter time constant of the speed feed forward device. Comprising a speed detector for outputting a speed calculation value implemented. As a result, the servo motor control device of the present invention can shorten the settling time without overshooting the position deviation even after the movement command is zero.

A servomotor control method according to the present invention is
A first step of generating a timing signal in the vicinity of a movement command that is a change in the position command becoming zero, and a position deviation that is a difference between the movement command and the position of the servo motor when the timing signal is input. Perform the integral calculation of
A second step of reducing the position deviation integral value and an integral calculation of a speed deviation which is a difference between the speed command obtained from the movement command and the speed of the servo motor when the timing signal is input are performed. And a third step of reducing the speed deviation integral value. This allows
The servomotor control method according to the present invention can shorten the settling time without overshooting the position deviation even after the movement command is zero.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION Preferred embodiments of a servomotor control device and control method according to the present invention will be described below with reference to the accompanying drawings.

<< Embodiment 1 >> FIG. 1 is a control block diagram of a servo motor controller according to Embodiment 1 of the present invention. As shown in FIG. 1, the position command θ * from the position command generator 4 is input to the servo driver 3 at regular intervals and arithmetically processed to control the input current to the servo motor 1. The servo driver 3 supplies a drive current to the servo motor 1 so as to match the position command θ * to perform position control. The servo motor 1 is provided with a position detector 2 for detecting the rotation angle of the servo motor 1. The position signal θ output from the position detector 2 is input to the servo driver 3 at regular intervals.

The servo driver 3 includes a position deviation calculator 5,
Position deviation proportional unit 6, position deviation integrator 7, speed feed forward calculator (hereinafter abbreviated as speed FF calculator)
8. Speed feed forward filter (hereinafter speed FF
(Abbreviated as filter) 9, speed calculator 10, speed deviation calculator 11, speed deviation proportional unit 12, speed deviation integrator 13,
A current command calculator 14, a current controller 15, and a timing generator 17 are included. In FIG. 1, Kpp in the position deviation proportionalizer 6 is a position proportional gain, Kpi in the position deviation integrator 7 is a position integration gain, Kff in the speed FF calculator 8 is a speed feedforward gain, and Kvp in the speed deviation proportionalizer 12 is a speed. Proportional gain, Kvi in velocity deviation integrator 13 is velocity integral gain, and s
Is the Laplace operator.

Next, the operation of the servo motor control apparatus according to the first embodiment will be described with reference to FIGS. FIG. 2 is a flowchart showing an operation in the servo driver 3 of the control device shown in FIG. 1, and shows a calculation process for one cycle in which the position loop process and the velocity loop process are executed.
The servo motor 1 is provided with a position detector 2 such as an optical encoder for detecting the rotation angle of the servo motor 1. The position signal θ output from the position detector 2 is read by the servo driver 3 at regular intervals. Further, the position command θ * from the position command generator 4 is input to the servo driver 3 at regular intervals. In the arithmetic processing of the servo driver 3, the position command θ * is read in processing step 1. The movement command as an input command to the position command generator 4 is a position change amount in a certain fixed period and is given by the sequencer. This sequencer outputs a movement command as a pulse, and in the position command generator 4, a pulse indicating the movement command is added by a counter and changed to a position command θ *. The servo driver 3 performs arithmetic processing corresponding to the position command θ * of the position command generator 4,
The drive current of the servo motor 1 is formed and supplied to control the position of the servo motor 1.

Next, the operation of the servo driver 3 in the first embodiment will be described. After the position command θ * output from the position command generator 4 is read in the processing step 1 of FIG. 2, whether or not the movement command is “0” is read based on the position command θ * as the first process in the timing generator 17. It is determined (processing step 2). Actually, as described above, the presence or absence of pulse input from the sequencer is determined. If the movement command is not "0", the flag F is set to "0" in processing step 6. Here, the movement command is the amount of position change in a certain period, so the current and latest position command is θ * (n), and the previously loaded position command is θ * (n-1).
Then, the movement command is defined by Δθ = θ * (n) −θ * (n-1).

First, the case of the normal movement operation in which the movement command is not "0" will be described. Position signal of position detector 2
And the position command θ * of the position command generator 4 are input to the position deviation calculator 5 and the calculation of θe = θ * −θ is performed for each position control cycle. The position deviation calculator 5 outputs the calculated position deviation θe to the position deviation proportional unit 6 and the position deviation integrator 7. The position deviation proportional device 6 calculates the product of the position deviation θe and the position proportional gain Kpp, and outputs the position deviation proportional value. Further, the position deviation integrator 7 calculates the product of the integrated value Σθe of the position deviation θe and the position integration gain Kpi, and outputs the position deviation integrated value.

In the first embodiment, in order to reduce the position deviation during operation, a velocity feedforward value obtained by differentiating the position command θ * is created, and the position deviation proportionalizer 6 is used.
And the position deviation integral value of the position deviation integrator 7 are added to form the speed command ω *. The position command θ * output from the position command generator 4 is a speed feedforward calculator 8 (hereinafter, the speed feedforward calculator 8 is abbreviated as a speed FF calculator 8 and other feedforwards are abbreviated as FF). Input). Speed FF
The computing unit 8 performs the differential processing of the backward difference and then the speed FF
The gain Kff is multiplied and the speed FF calculation value is output. Here, in order to reduce the position deviation θe during operation, the speed F
The F gain Kff takes a value of 1. When the resolution of the position command θ * is coarse, the speed FF calculation value is input to the speed FF filter 9, the primary low-pass filter (LPF) processing is performed, and a smooth speed FF value is output. However,
In the first embodiment, since the resolution of the position command is high, the speed F
The filter time constant of the F filter 9 is set to "0", and the filtering process is not performed.

As described above, the signals output from the position deviation proportional unit 6, the position deviation integrator 7, and the speed FF filter 9 are added in the speed deviation calculator 11, and the speed command ω * as an intermediate state quantity is added. It is calculated. That is, the current position deviation integrated value is Σθe (n), and the previous position deviation integrated value is Σθe.
Assuming that (n-1), the speed command ω * is obtained by executing the calculation of the following equation (4) every position control cycle. This process is performed in process step 7 and corresponds to the position loop process. In Expression (4), LPF indicates low-pass filter calculation processing, and in the first embodiment, if LPF = 1, no processing is performed.

[0031]

[Equation 4]

Next, the speed calculator 10 similarly converts the position signal θ output from the position detector 2 at regular intervals into speed by reverse differential processing and, if necessary, smoothes it by a filter to make the speed signal ω. Is output. In the first embodiment, the filter time constant of the speed calculator 10 is “0”. The speed signal (detected speed) ω output from the speed calculator 10 is input to the speed deviation calculator 11 as with the speed command ω *, and the calculation of ωe = ω * −ω is performed. Then, the speed deviation calculator 11 outputs the speed deviation ωe.

Next, the speed deviation ωe output from the speed deviation calculator 11 is input to the speed deviation proportionalizer 12 and the speed deviation integrator 13. The speed deviation proportionalizer 12 determines the position deviation ωe
And a velocity deviation proportional amount, which is the product of the velocity proportional gain Kvp and the calculated value, is output. In addition, the speed deviation integrator 13 calculates the current speed deviation integrated value by Σωe (n) and the previous speed deviation integrated value by Σω
Assuming e (n-1), the following equation (5) is calculated to output the speed deviation integrated value Σωe (n) (processing step 8 in FIG. 2).

[0034]

[Equation 5]

Then, the calculated speed deviation proportional amount and speed deviation integrated amount are added by the current command calculator 14 and output as the current command value I * (process step 9 in FIG. 2). The current command calculator 14 performs the calculation represented by the following equation (6) to calculate the current command value I *.

[0036]

[Equation 6]

The current command calculated as described above is input to the current controller 15 (processing step 10 in FIG. 2) and controlled by the current minor so that the servo motor 1 is supplied with the current of the current command value I *. It As described above, the current command is output from the speed command ω * and the detected speed ω, and this processing corresponds to the speed loop processing.

As long as a position command is input and a movement command is issued, processing step 1 → processing step 2 is performed every processing cycle.
→ processing step 6 → processing step 7 → processing step 8 →
The position loop processing and the speed loop processing of processing step 9 → processing step 10 are repeatedly executed. In this way, the position loop processing and the speed loop processing are repeatedly executed, and the position and speed of the servo motor 1 are feedback-controlled. In the above processing, the timing generator 17 determines whether or not there is a movement command in processing step 2,
If there is no movement command and it is determined that the movement command is "0", it is determined in step 3 whether the flag F is "1".
Since the flag F is initially set to "0" in processing step 6, processing steps 3 to 4 are processed.
Move to.

Position deviation θ read by the position loop processing
e is compared in step 4 with a preset threshold value ε. The absolute value of the position deviation θe is the threshold width ε
When the value is larger, the timing generator 17 does not output the timing signal, and the normal position integration processing is performed in the processing step 7 and the velocity integration processing is performed in the processing step 8. Then, the current command value I * is calculated in processing step 9, and this current command value I * is sent to the current controller 15 in processing step 10. That is, when the absolute value of the position deviation θe is larger than the threshold width ε, normal speed loop processing is executed. Therefore, as long as the absolute value of the position deviation θe is larger than the threshold value ε, processing step 1 → processing step 2 → processing step 3 → processing step 4 → processing step 7
→ process step 8 → process step 9 → process step 10
The process of is repeated.

On the other hand, in processing step 4, the position deviation θ
If the absolute value of e is equal to or smaller than the threshold value ε, the process proceeds to processing step 5. This case corresponds to the case where the timing signal is output from the timing generator 17. That is, when the timing signal from the timing generator 17 is input to the position deviation integrator 7, the position deviation integrated value Σθe up to the previous cycle is processed in processing step 5.
(N-1) is multiplied by the setting coefficient k2 to calculate the position deviation integrated value Σθe (n) of the cycle, and the speed deviation integrated value Σωe (n-1) up to the previous cycle is multiplied by the setting coefficient k3. The speed deviation integrated value Σωe (n) of the period is calculated. Values of 0 and less than 1 are set for the coefficients k2 and k3. Since the coefficients k2 and k3 are set to 0 in the first embodiment, the position deviation integrated value and the speed deviation integrated value are 0. The arithmetic expression in processing step 5 is shown in the following expression (7).

[0041]

[Equation 7]

In processing step 5, the flag F is set to "1".
After setting to, processing step 9 and processing step 10
Is carried out. As described above, since the flag F is set to "1" in the processing step 5, even if the movement command is "0" in the next control loop, the processing shifts from the processing step 3 to the processing step 7. In this way, in the control loop, the position deviation integrated value Σθe (n) and the speed deviation integrated value Σωe (n) are switched to small values only once in processing step 5, so the current command value I * becomes small and the servo motor The output torque of 1 also becomes small. As a result, the occurrence of overshoot of the position deviation after the movement command becomes "0" is prevented.

Next, when the viscous friction of the controlled object is small,
The effectiveness of the configuration and the operation of the control device according to the first embodiment configured as described above will be described with reference to FIGS. FIG. 3 shows that, in the control device according to the first embodiment, the value of the position deviation integrator 7 is set to "0" when the movement command is "0" (when the movement command is zero) with respect to the control object having small viscous friction. The differential value (speed command) (a) and the position deviation (b) of the position command are shown. Here, since the velocity integration gain of the velocity deviation integrator 13 is "0", only the position integration operation is calculated as the integrator. The position deviation proportionalizer 7 and the speed deviation proportionalizer 12 operate. In FIG. 3, the position deviation as shown in (b) is obtained with respect to the speed command which is the differential value of the position command as shown in (a). As shown in (a) and (b) of FIG.
Large positional deviations occur at points B, C, and the positional deviation integrator 13 causes the positional deviations to gradually converge to zero. That is, A point → B point, B point → C
When approaching the point, point C → point D, the position deviation becomes close to “0”.

As shown in FIG. 3B, the output of the position deviation integrator 7 is reduced when the movement command at the final point D is zero, so that the position deviation can be substantially changed at the point D. At the point D in FIG. 3B, the position deviation changes to “0”. As shown in FIG. 3, when the movement command is zero, there are before movement (before point A) and after movement command completion (after point D), but unless otherwise specified, in the following embodiment, the movement command is It means the completed point D. Therefore, in the first embodiment, the vicinity of the movement command being “0” means that the vehicle is decelerating and the movement command becomes near zero before and after the point D including the point D (time ± number of the point D).
msec) represents the time range.

FIG. 4 is shown as a comparison of FIG. 3, in which the speed command (a) which is the position command differential value and the position deviation (b) when the output of the position deviation integrator 7 is not reduced when the movement command is zero. ) Respectively. It can be seen that the position deviation occurs and the convergence of the position deviation is delayed after the point D when the movement command is zero. Further, when the viscous friction is small, it is sufficient to provide either the speed deviation integrator 13 or the position deviation integrator 7, but when the viscous friction is large, the position deviation integrator 7 and the speed deviation integrator are integrated. The necessity of using the container 13 together will be described below. There is no position deviation integrator 7 during acceleration shown in the periods A to B of FIG. 3 and deceleration shown in the periods C to D of FIG.
If only works, the following formula (8)
The position deviation θe is calculated at.

[0046]

[Equation 8]

In the equation (8), D is a viscous friction coefficient, K
Is the acceleration constant. On the contrary, when the speed deviation integrator 13 is not provided during acceleration / deceleration and only the position deviation integrator 7 acts as the integrator, the position deviation θe is similarly obtained by the following equation (9) using the final value theorem.

[0048]

[Equation 9]

However, when the position deviation integrator 7 and the speed deviation integrator 13 are used together, a steady deviation does not occur even during acceleration / deceleration. FIG. 5 shows that for a control object with large viscous friction,
The speed command (a) and the position deviation (b) which are the position command differential values when the position deviation integrator 7 and the speed deviation integrator 13 are used as the integrator are shown. As shown in (b) of FIG. 5, the position deviation is “0” even during acceleration and deceleration, and it can be confirmed that the position deviation is set to almost 0 even when the movement command is zero.

FIG. 6 is shown as a comparison with FIG. FIG. 6 shows the speed command (a) and the position deviation (b) which are the position command differential values when the control target having a large viscous friction is controlled by only the speed deviation integrator 13. As shown in FIG. 6, it can be confirmed that the position deviation does not return to “0” at the time of acceleration or deceleration, but has a constant position deviation. As described above, by using the position integral term and the velocity integral term together for the controlled object having a large viscous friction, the position deviation converges to near zero in the vicinity where the movement command becomes “0”. . Further, the position deviation overshoot is prevented by reducing the position deviation integrated value and the speed deviation integrated value in the vicinity of the completion of the position command.

In recent years, in a manufacturing apparatus using a servo motor, the settling time from the completion of the position command to the setting within a predetermined settling width is required to be about several msec, and the settling width is very large. It is a narrow one. In an actual machine, the settling width is set to a few microns or less.
Therefore, a slight deviation of the position deviation after the movement command is zero becomes a large difference because the settling time becomes long when viewed microscopically. Therefore, when the position command is completed, the position deviation is kept within the settling width, so that the settling is performed at the same time as the position command is completed, and the settling time can be significantly shortened.
The timing of outputting the timing signal from the timing generator 17 is ideal at the point D where the movement command becomes “0” in the ideal case where the filter delay and the like can be ignored. However, in an actual machine, there is a slight calculation delay, and in that case, the same effect can be obtained by resetting the integral value before and after point D.

Actually, the calculation of the position loop processing and the speed loop processing including various filter processings and the processing by the timing generator 17 are realized by software and implemented by a microcomputer. Therefore, the first embodiment has been described as a configuration in which various conversion coefficients including the torque constant are included in each gain to directly obtain the current command value. However, even if the step of obtaining the current command by dividing the torque command by the torque constant after obtaining the torque command is separated from other configurations as one element, the first embodiment
It goes without saying that it has the same effect as that of the above configuration. Further, although the configuration in which the speed FF filter 9 is arranged after the speed FF calculator 8 is described in the first embodiment, the present invention is not limited to this structure, and the speed FF calculator 8 and the speed FF are not limited to this structure. It goes without saying that the order of the filters 9 may be reversed.

In the present invention, the velocity FF calculator 8 and the velocity FF filter 9 constitute a velocity feed forward device, which is referred to as a velocity feed forward device in the claims. Further, the movement command in the first embodiment is obtained by performing a difference calculation on the position command of the position command generator 4, but the same effect is obtained even if the pulse signal itself coming from the sequencer is used as the movement command. Needless to say. The control structure of the present invention is not limited to the control structure in which the velocity loop is arranged inside the position loop shown in FIG. It goes without saying that the same effect can be obtained even with a PID (proportional, integral, derivative) control configuration for position and other control configurations. Further, in the timing generator 17 of the first embodiment, the processing step 2, the processing step 3 shown in FIG.
And the judgment processing is executed in the processing step 4, but the judgment processing in the processing step 4 is omitted, or before and after the movement command becomes “0” in the processing step 2 as described in the second embodiment to be described later. It goes without saying that the same effect can be obtained even if the configuration is such that the timing signal is output when (the movement command is near zero), and this configuration is also included in the present invention.

Further, in the processing step 5, the example in which the coefficients k2 and k3 are set to 0 in the first embodiment has been described. Alternatively, a slightly larger value such as 0.7 may be given. When the signal is output from the timing generator 17, the processing step 5 is changed to be executed several times instead of once, so that the position deviation integrator 7 and the speed deviation integrator 13 are accumulated. Value decreases gradually. As a result, when the damping of the overall control system is small (the stability margin is small), it is possible to prevent the position deviation from fluctuating due to a sudden change of the input. Therefore, according to such a configuration, the position deviation converges smoothly without vibration, and the settling time can be shortened even when the damping of the system is low. Here, in order to gradually reduce the integrated value, the calculation may be performed by the low-pass filter given by the following equation (10).

[0055]

[Equation 10]

In equation (10), Σθe_f is the final target value, and in the equation on the first line in equation (7), Σθe_f is 0.
Can be regarded as the case. Further, assuming that the position deviation integral value before Σθed is decreased, the following expression (1
The same effect can be obtained by linearly decreasing as in 1).

[0057]

[Equation 11]

<Second Embodiment> Next, a servo motor control device according to a second embodiment of the present invention will be described with reference to the drawings. First, the function of the servo motor control device according to the second embodiment will be briefly described. In the above-described first embodiment, the time constants of the filters of the speed FF filter 9 and the speed calculator 10 are set to 0 (the same as when there is no filter). In practice, the time constant of the filter may be used as "0" in this way, but the position command is smoothed when the resolution of the position command input (pulse) or the actual position output (pulse) is low. Therefore, processing by the speed FF filter 9 is performed, and filter processing is performed by the speed calculator 10 to smooth the position output.
When the filtering process is performed in this way, a position deviation may occur when the movement command is zero. The servomotor control device according to the second embodiment is a device that eliminates the occurrence of the position deviation as described above and changes the position deviation within the settling width to shorten the settling time.

FIG. 7 shows a control block diagram in the servo motor control apparatus according to the second embodiment. In FIG. 7, components having the same functions and configurations as those of the control device of the first embodiment described above are designated by the same reference numerals, and the description thereof will be omitted. The servo driver 23 in the second embodiment is similar to the first embodiment described above in that the position deviation calculator 5, the position deviation proportionalizer 6,
Position deviation integrator 7, speed feed forward calculator (hereinafter abbreviated as speed FF calculator) 8, speed calculator 10,
It has a speed deviation calculator 11, a speed deviation proportionalizer 12, and a current controller 15. In the servo motor control device according to the second embodiment, a configuration different from that of the first embodiment is a speed feedforward filter (hereinafter, abbreviated as a speed FF filter) 29 and a timing generator 37 in the servo driver 23. The integrator 13 and the current command calculator 14 are not provided. In FIG. 7, Kpp in the position deviation proportionalizer 6 is position proportional gain, Kpi in the position deviation integrator 7 is position integration gain, Kff in the speed FF calculator 8 is speed feed forward gain, and Kvp in the speed deviation proportionalizer 12 is speed. Proportional gain, and s is the Laplace operator.

Next, the operation of the servo motor controller according to the second embodiment will be described with reference to FIGS. 7 and 8. FIG. 8 is a flowchart showing the operation in the servo driver 23 of the control device shown in FIG. 7, and shows the calculation processing for one cycle in which the position loop processing and the speed loop processing are executed. The servo motor 1 is provided with a position detector 2 such as an optical encoder for detecting the rotation angle of the servo motor 1. The position signal θ output from the position detector 2 is read by the servo driver 23 at regular intervals. A position command θ * is input to the servo driver 23 from the position command generator 4 at regular intervals. Servo driver 2
In the calculation of 3, the position command θ * is read in processing step 1. The servo driver 23 performs arithmetic processing corresponding to the position command θ * of the position command generator 4, forms and supplies the drive current of the servo motor 1, and controls the position of the servo motor 1.

Next, the operation of the servo driver 23 in the second embodiment will be described. Processing step 1 in FIG.
Position command θ * output from position command generator 4 at
After being read, the timing generator 37 determines whether or not the movement command is “0” based on the position command θ * (processing step 2). If the movement command is not "0", the flag F is set to "0" in processing step 6. Here, the movement command is the amount of position change in a certain period, so if the current latest position command is θ * (n) and the previously loaded position command is θ * (n-1), the movement command is Δ
It is defined by θ = θ * (n) −θ * (n-1).

First, the case of the normal movement operation in which the movement command is not "0" will be described. Position signal of position detector 2
And the position command θ * of the position command generator 4 are input to the position deviation calculator 5 and the calculation of θe = θ * −θ is performed for each position control cycle. The position deviation calculator 5 outputs the calculated position deviation θe to the position deviation proportional unit 6 and the position deviation integrator 7. The position deviation proportional device 6 calculates the product of the position deviation θe and the position proportional gain Kpp, and outputs the position deviation proportional value. Further, the position deviation integrator 7 calculates the product of the integrated value Σθe of the position deviation θe and the position integration gain Kpi, and outputs the position deviation integrated value.

The position command θ * output from the position command generator 4 is input to the speed FF calculator 8. The speed FF calculator 8 performs the differential processing of the backward difference and then multiplies the speed FF gain Kff to output the speed FF calculation value ωff (n).
Since the resolution of the position command in the second embodiment is coarse, the speed FF calculation value is input to the speed FF filter 29. If the resolution of the position command is rough, the current command (torque) becomes noisy, and a sound is generated from the control target. In the speed FF filter 29 of the second embodiment, the first-order low-pass filter (LPF) represented by the following expression (12) is used.
Is performed, and a smooth speed FF calculation value ωff_lpf (n) is output.

[0064]

[Equation 12]

In the equation (12), a is a constant. As described above, in the second embodiment, the position deviation proportionalizer 6
The added value of the outputs from the position deviation integrator 7 and the speed FF filter 29 becomes the speed command ω *. That is, the position loop calculation of the following equation (13) is performed for each position control, and the speed command ω * is output (processing step 7).

[0066]

[Equation 13]

Next, the speed calculator 10 calculates the position signal θ output from the position detector 2 at regular intervals as shown in the following equation (14) by the reverse differential processing as in the first embodiment. To the velocity signal ω
Output (n).

[0068]

[Equation 14]

In the equation (14), b is a constant. Further, similarly to the first embodiment, the speed signal ω output from the speed calculator 10 is input to the speed deviation calculator 11, and ωe
= Ω * −ω is calculated and the speed deviation ωe is output.

Next, the speed deviation proportionalizer 12 determines the speed deviation ωe
Then, the product of the speed proportional gain Kvp and the speed proportional gain Kvp is calculated by the following equation (15) to output the current command value I *.

[0071]

[Equation 15]

The arithmetic processing in the processing step 8 shown in FIG. 8 implements the equations (14) and (15). The resulting current command is input to the current controller 15 (processing step 9). The current controller 15 is the servo motor 1
The servo motor 1 is driven and controlled by supplying the current of the current command value I * to the.

As long as a position command is input and a movement command is issued, processing step 1 → processing step 2 is performed every processing cycle.
→ processing step 6 → processing step 7 → processing step 8 →
The position loop processing and the speed loop processing of processing step 9 are repeatedly executed. In this way, the position loop processing and the speed loop processing are repeatedly executed, and the position and speed of the servo motor 1 in the second embodiment are feedback-controlled. In the above process, the timing generator 37 determines the presence / absence of the movement command in the processing step 2. If the movement command disappears and the movement command is determined to be "0", it is determined in step 3 whether the flag F is "1". To judge. Since the flag F is initially set to "0" in processing step 6, the processing shifts from processing step 3 to processing step 4.

Position deviation θ read by the position loop processing
It is determined whether or not e is less than or equal to the threshold value ε preset in processing step 4. When the absolute value of the position deviation θe is larger than the threshold width ε, the normal position integration processing is performed in the processing step 7 corresponding to the case where the timing signal is not output from the timing generator 37. Then, the current command value I * is calculated in processing step 8, and this current command value I * is sent to the current controller 15 in processing step 9. That is, when the absolute value of the position deviation θe is larger than the threshold width ε, normal position loop processing and speed loop processing are executed. Therefore, the absolute value of the position deviation θe is
As long as it is larger, the processing of processing step 1 → processing step 2 → processing step 3 → processing step 4 → processing step 7 → processing step 8 → processing step 9 is repeated.

On the other hand, in processing step 4, the position deviation θ
If the absolute value of e is equal to or smaller than the threshold value ε, the process proceeds to processing step 5. That is, the timing generator 37 outputs the timing signal only when executing the processing step 5. When the timing signal from the timing generator 37 is input to the position deviation integrator 7,
In processing step 5, the position deviation integrated value Σθe (n-1) up to the previous cycle is multiplied by the setting coefficient k2 to calculate the position deviation integrated value Σθe (n) of the cycle, and the speed FF calculation value is set by the setting coefficient k4. And the speed FF calculation value ωff for the relevant cycle
Calculate (n). Values of 0 and less than 1 are set for the coefficients k2 and k4, and the position deviation integrated value and the speed FF are set.
The calculated value is a small value. The following expression (16) is an arithmetic expression in processing step 5.

[0076]

[Equation 16]

Finally, as shown in equation (16), the flag F
Is set to "1". Then, processing step 8 and processing step 9 are performed. As described above, since the flag F is set to "1" in the processing step 5,
In the next control loop, even if the movement command is "0", the process shifts from the processing step 3 to the processing step 7. Thus, in the control loop, processing step 5 is performed only once.
Since the position deviation integrated value Σθe (n) and the speed FF calculated value are switched to small values, the current command value I * becomes small and the output torque of the servomotor 1 also becomes small. As a result, the occurrence of overshoot of the position deviation after the movement command becomes "0" is prevented.

When the above-mentioned processing operation is carried out, when the filter processing is executed, a position deviation similar to the position deviation waveform shown in FIG. 3 is obtained. That is, the position deviation is generated from each of the points A, B, and C when the speed changes suddenly as shown in FIG. 3B. However, the position deviation integrator 13 works to gradually move each position deviation. Has converged to "0". Then, when the movement command at the point D in FIG. 3B is zero, the outputs of the position deviation integrator 7 and the speed FF filter 29 are decreased (set to 0 in FIG. 3), so that the position deviation is continued after the point D. It is possible to change the position deviation in the vicinity of zero without causing the fluctuation of. When the velocity FF calculation value is set to "0" once after the movement command is zero, it remains "0" thereafter, as is apparent from the second equation of the above equation (13).

FIG. 9 is a waveform diagram as a comparison with FIG. 3, in which the speed command (a) and the position deviation (b) which are position command differential values when the output of the speed FF filter 29 is not reduced.
Are shown respectively. As shown in FIG. 9B, it can be confirmed that a position deviation occurs after the movement command is zero. The reason for this is that when the speed FF filter 29 performs the filter calculation and outputs the final speed FF calculated value, the speed FF calculated value becomes "0" due to the filter delay even when the movement command is zero. However, a positive value has occurred. As a result, the actual position overshoots the position command, and a negative position deviation (θe = θ * −θ) overshoot occurs.

That is, when the speed FF filter 29 operates, the position deviation integral value and the speed FF calculation value from the speed FF filter 29 are reduced so that the position deviation does not overshoot after the movement command becomes zero. , The position deviation converges near zero as it is. In this way, since the position deviation continues to stay within the settling width, the settling becomes simultaneous with the completion of the position command, and the settling time can be greatly shortened. In Embodiment 2 described above, the speed FF filter 29 and the filter of the speed calculator 10 are first-order low-pass filters, but the present invention is not limited to such a configuration, and the order is different. Low-pass filter (II
R filter) and moving average filter (FIR filter)
Needless to say, such as may be used.

In the second embodiment, the speed deviation integrator 13 and the current command calculator 1 described in the first embodiment are used.
Although the case where No. 4 is not provided has been described, it goes without saying that the same effect as that of the first embodiment can be obtained by providing the speed deviation integrator 13 and the current command calculator 14 as in the first embodiment. . In the flowchart shown in FIG. 8, the processing of processing step 4 can be carried out even if it is omitted, and the processing of processing step 5 should be executed only once when the movement command becomes “0” in processing step 2. The same effect can be obtained.

If a torque filter or the like is inserted, the position deviation integrated value output from the position deviation integrator 7 or the speed FF filter 2 will be described below.
By decreasing either one of the speed FF calculation values output from 9 according to the timing signal from the timing generator 27, it is possible to prevent the deviation of the position deviation after the movement command is zero. Various methods for preventing the deviation of the position deviation after the movement command is zero will be described below. First, a method of controlling the deviation of the position deviation by decreasing the position deviation integral value before the movement command becomes zero will be described. Even if the control target has high rigidity, if the calculation delay of the position loop or the speed loop is large or if the torque filter (this torque filter will be described in the third embodiment) is large, the movement command becomes zero due to these delays. After becoming
Position deviation fluctuations occur. A method of preventing the deviation of the position deviation and shortening the settling time will be described below.

During the deceleration period from point C to point D shown in FIG. 9, the servo motor 1 is decelerating and negative torque is generated. Therefore, when the calculation delay is large or the torque filter is large, even if the position deviation integral term is reduced to "0" at the point D, the negative torque does not immediately become "0" due to the delay, and after a while after the point D. Negative torque is applied for a period of time.
As a result, the positional deviation occurs in the positive direction. Therefore, when the position deviation integrated value is reduced before the movement command is zero (point D) in anticipation of these delays, the torque becomes near zero when the movement command is zero. The timing for decreasing the position deviation integrated value before the point D depends on the delay amount at that time, but generally a value within 0.1 msec to 2 msec is taken. In this way, when the delay is large, the deviation of the position deviation can be reduced and the settling time can be shortened by making the timing of decreasing the position deviation integral term earlier than when the movement command is zero.

Similarly, the speed FF calculation value produces a positive torque. Therefore, when the delay of the negative torque is large by the above filter, the timing signal for reducing the speed FF calculator is delayed to give a positive input for canceling the negative input to reduce the deviation of the position deviation and shorten the settling time. be able to. In this case, position deviation integrated value and speed FF
It is necessary to separately provide a timing signal for reducing the calculated value. The details will be described later.

On the other hand, when the rigidity of the controlled object is low and the gain cannot be set high, not only positive but also negative position deviation may occur when the movement command is zero. Therefore, first, the timing generator 37 calculates the position deviation integrated value output from the position deviation integrator 7 and the speed FF calculation value output from the speed FF filter 29 so as to cancel the position deviation occurring when the movement command is zero. The output of the timing signal to be decreased is set simultaneously or individually at time points before and after the time when the movement command is zero. This reduces the position deviation that occurs when the movement command is zero and shortens the settling time.

First, the case where the timing signal from the timing generator 37 is simultaneously output to the position deviation integrator 7 and the speed FF filter 29 will be described. FIG. 10 is a waveform diagram showing the velocity FF calculation value and the position deviation integrated value near the movement command zero (point D). When the speed FF calculated value and the position deviation integrated value are set to “0” at the point E earlier than the point D, the positive value of the speed FF calculated value having an absolute value larger than the position deviation integrated value is “0”. Therefore, the change in the speed FF calculation value is calculated by the speed minor and is equivalently E
A response similar to that given a negative input torque at a point is obtained. That is, at this time, the position deviation occurs in the positive direction.

Next, consider the case where the position deviation integral value is set to zero at point F at a time later than point D (movement command zero). At point F, the speed FF calculation value is already “0”. Therefore, the timing for setting the position deviation integrated value to "0" is delayed, and the position deviation occurs in the positive direction due to the negative torque as at point E. When the movement command is 0, the position deviation is 0, and the external force such as the gravitational term does not act, the required current command is 0. The velocity FF calculation value that exerts an influence as a result causes a position deviation. That is, although the response differs depending on the position deviation integrated amount and the value of the speed FF calculation value,
It can be easily understood that it is possible to control the position deviation in the positive direction by changing the timing signal to a point.

Next, a case will be described in which the timing deviation reducing value and the timing FF calculation value are individually given. First, consider a case where the position deviation integrated value is set to “0” at point E in FIG. 10 and the speed FF calculation value is set to “0” at point G in FIG. 10 (operation A). In this case, even after the position deviation integrated value becomes “0” at point E, the position deviation is generated in the negative direction by continuously applying the positive torque based on the positive speed FF calculation value. Conversely, when the position deviation integrated value is set to “0” at the G point and the speed FF calculated value is set to “0” at the E point, a negative value is obtained even after the speed FF calculated value becomes “0” at the E point. The position deviation is generated in the negative direction by continuing to add the position deviation integrated value. That is, the position deviation integrated value is reduced based on the position deviation near the movement command zero time (point D) by individually providing the timing signal for decreasing the position deviation integrated value and the speed FF calculation value. The direction and amount of the position deviation in the vicinity of the movement command being “0” can be controlled by individually controlling the timing of the movement and the timing of decreasing the speed FF calculation value.

In the second embodiment, when the timing generator 37 determines the E point before the movement command becomes "0", the movement command is decreased and the movement command becomes less than a certain set value. Time is judged to be E point. Further, in the case of the G point, the timing generator 37 determines as the G point after a predetermined period of time has elapsed after the movement command becomes “0”, and at that time, the timing generator 37 outputs the timing. It suffices to output a signal. In the second embodiment,
The timing generator 37 is configured to output a timing signal to each of the position deviation integrator 7 and the speed FF filter 29. In the above, when there is a position deviation near the movement command zero (near movement command completion),
When the direction of the position deviation is controlled by the timing signal from the timing generator 37, the following phenomenon occurs with respect to the position deviation.

When there is a position deviation near the time when the movement command is zero, when the position deviation is converged to "0", the integration operation by the position deviation integrator 7 is performed in a slight period during which the position deviation occurs. Be seen. When the integration operation is performed in this manner, the position deviation integrator 7 overshoots by the integration operation. Therefore, when the position deviation occurs near the time when the movement command is zero, the position deviation integrator 7 is set not to operate for a preset period (operation B). This makes it possible to reduce the occurrence of overshoot. As an operation close to the above-described operation in which the operation A and the operation B are combined, the point E is a time point when the decreasing movement command becomes equal to or less than a preset value, and the preset ts1 period from the point E. The position deviation integrated value is set to "0" only for (the period from point E to point G). In addition, the speed FF calculation value is an operation that is set to "0" only once after the period of ts2 has elapsed from the point D when the movement command became 0 (point H in FIG. 14, which will be described later). Explained.

FIG. 11 shows a position for executing the above contents.
It is a flow chart of speed. In the second embodiment,
F1 is a flag for setting the flag F1 to 1 when the operation of setting the position integration value to "0" is started and for continuing the operation for the ts1 period even when the movement command becomes 0 during the operation. F2 sets the flag F2 to "0" before executing the speed FF calculation value "0", and after executing it, sets the flag F2 to "1", so that the movement amount is "0". This is a flag for executing the speed FF calculation value "0" only once. t1 is a timer that starts counting after starting the operation of setting the position integral value to “0”. In the second embodiment, the period t1 in which the position integral value is “0” is managed by comparing it with ts1. ing. t2 is
It is a timer that starts counting after the movement amount becomes “0”, and in the second embodiment, it is managed by comparing t2 with ts2. ts1 is a preset period in which the position integration value is held at zero. ts2 is a preset period from when the movement amount becomes “0” to when the speed FF calculation value becomes “0”. In processing step 51 of FIG. 11, initial values of F1 = 0 and t1 = 0 are set, and in processing step 52, a position command is read. Next, in process step 53, the rotation direction of the servomotor 1 is determined, which is clockwise or counterclockwise. Also, Δθ (n) = θ
* (n) -θ * (n-1) is performed to calculate the movement command (speed).
By this calculation, if Δθ (n) −Δθ (n-1) <0,
It is determined that the vehicle is decelerating, and whether the position movement amount is equal to or less than the preset movement amount Δθs is determined by Δθ (n) <Δθs. Then, if the respective conditions are not satisfied (“No”, which corresponds to the signal N), the process proceeds to the processing step 54. In processing step 54, F1 =
It is determined whether or not 1. F1 in processing step 51
Since = 0 is still set, the process shifts from the processing step 54 to the processing step 58, the position deviation integral calculation is performed, and the position deviation integral value is output.

Next, in processing step 61 of FIG. 11, it is judged whether or not there is a movement command (Δθ = 0), and “NO”.
If so, after setting F3 = 0 and t2 = 0 in the processing step 62, the speed FF calculation value (ω
ff (n)) is output. This speed FF calculation value (ωff
The calculation of (n)) is the same as a part of the calculation in the processing step 7 of FIG. 8 described above. Next, in processing step 68, the calculation processing of the speed command (ω *) is performed in the same manner as part of the calculation in processing step 7 of FIG. Then, in processing step 69, the speed minor loop calculation is carried out in the same manner as in the processing step 8 of FIG. 8 to obtain the current command value I *. In processing step 70, the calculated current command value I * is output to the current controller 15, and the current controller 15 controls the current supplied to the servomotor 1. The above processing steps are repeated.

On the other hand, in the processing step 53 of FIG. 11, when all the conditions are satisfied (“affirmative” and corresponds to the symbol Y), in the processing step 55, F1 =
1 is set, and in the process step 56, the period t1 in which the position deviation integrated value is set to "0" is preset to ts1.
Determine if the period has expired. If t1 ≧ ts1 is not satisfied (“negative”), k2 = in processing step 59.
0, and the position deviation integral value is set to 0. The timing generator 37 outputs a timing signal to the position deviation integrator 7 when the processing of the processing step 59 is performed for the first time.

Next, at processing step 60, time t1
Is added to the additional time tp. And, as mentioned above,
Process step 61 → process step 62 → process step 6
The processing of 6 → processing step 68 → processing step 69 → processing step 70 is performed. The above processing step 53 → processing step 55 → processing step 56 → processing step 59
→ The processing loop from processing step 60 to processing step 70 is executed multiple times.

Then, in processing step 53, when the movement command is "0" and Δθ (n) -Δθ (n-1) <0 is "negative", the processing shifts to processing step 54. In the processing step 54, since the flag F1 is still set to 1 in the previous processing step 55, the processing loop from processing step 56 → processing step 59 → processing step 60 to processing step 70 is executed. At this time,
The position deviation integrated value remains "0". afterwards,
When the set time ts1 has elapsed, the process proceeds to the process step 57 by the judgment in the process step 56, and F1, t1
Are set to "0". When F1 is set to "0", the determination at processing step 54 becomes "affirmative" in the next cycle, and the position deviation integration operation is performed at processing step 58. As described above, the position deviation integral value continues to be zero only during the ts1 period after the condition of the processing step 53 becomes "affirmative".

Next, the loop for calculating the speed FF calculation value will be described. When the movement command is judged to be “0” in the processing step 61 of FIG. 11, the processing shifts to the processing step 63. In the processing step 63, the flag F3 is still set to "0" in the previous processing step 62, so that the processing proceeds to the processing step 64. In processing step 64, time t
2 is compared with a preset set time ts2, and the time t
If 2 has not reached the set time ts2, the additional time tp is added to the time t2 in the processing step 65. Then, in processing step 66, the speed FF calculation value is calculated and output. On the other hand, when the time t2 is equal to or exceeds the set time ts2 in the processing step 64, the speed FF calculation value is set to "0" in the processing step 67 (in the calculation formula in the processing step 67,
k4 = 0). In the processing step 67, F2 =
Set to 1. The timing generator 37 outputs a timing signal to the speed FF filter 29 when performing the processing of the processing step 67. Therefore, “Yes” is given in the processing step 63 of the next control cycle, and the speed FF calculation processing is executed again in the processing step 66.
That is, the process of the process step 67 is performed only once.
In the second embodiment, the speed FF calculated value is already "0" when the decrease timing signal of the speed FF calculated value is a little slower. Obviously, there is no need to output timing signals from 37.

In the above-described processing, the position deviation integrated value can be set to "0" for a certain period or the speed FF calculation value can be set to "0" at a certain time point according to the conditional expression including the set constant. In the above processing, the set value Δθ in the processing steps 53, 56, 61, 64 including the setting constant
There are two methods of setting s, ts1, Δθ, and ts2. One is a method of setting a person by seeing the response. The other is a method of setting a value by a constant setting means configured in the timing generator 37. The constant setting means automatically outputs the position deviation from the timing generator 37 according to the position deviation amount and the deviation direction in the vicinity of the movement command zero (including the movement command zero) based on the timing signal and the position deviation generation direction. The constant of the conditional expression for outputting the timing signal that reduces the values of the integrator 7 and the speed FF filter 29 is determined. The positional relationship and the timing relationship for decreasing each value should be quantitatively recognized in advance through experiments or the like. As described above, by providing the constant setting means in the timing generator 37, it is possible to automatically reduce the position deviation near the movement command zero.

FIG. 12 shows a speed command (a) and a position deviation (b) which are differential values of the position command of the control device of the present invention based on the flowchart of FIG. 11 when the gain is small.
FIG. 13 is an enlarged view of the vicinity of point D of the position deviation waveform of FIG. FIG. 14 is a waveform diagram showing the position deviation integrated value and the speed FF calculation value near the movement command zero. Figure 15
Is the position deviation integrated value and velocity F at point D when the position command is zero.
The speed command (a) and the position deviation (b) which are the differential values of the position command when the F operation value is set to "0" only once are shown. Figure 1
6 is an enlarged view showing the vicinity of point D of the waveform of the positional deviation in FIG. 13 and 16 have the same time scale on the horizontal axis. By comparing the waveform diagrams of FIG. 13 and FIG. 16, it can be confirmed that the waveform of FIG. 13 has the positional deviation within the settling width in a shorter settling time. In the waveform of FIG. 16, the settling time is extended because the convergence of the position deviation within the settling width depends on the value accumulated in the position deviation integrator 7 little by little.
Further, as shown in FIG. 14, it can be confirmed that the position deviation integrated value is “0” from the E point to the G point, and the speed FF calculation value is zero at the H point. The position deviation integrated value shown in FIG. 14 is almost zero even after the point G, because the occurrence of position deviation is small.

FIG. 17 shows the case where the speed FF calculation value is reduced to "0" only once at the H point and the position deviation integrated value is reduced to "0" only once at the E point. It is an enlarged view of a position deviation. FIG. 18 shows the position deviation integrated value and the speed FF calculated value shown in FIG. The overshoot of the position deviation in FIG. 17 is slightly larger than that in FIG. Further, as is clear from FIG. 18, as a cause thereof, it can be confirmed that the position deviation integrated value is generated in the positive direction from the point E. That is, the input torque in the positive direction is applied and the overshoot becomes large. 13 and 17
Comparing the above, there is no difference in the effect of setting the settling width under this condition, but if the setting of the settling width is made narrower, in the case shown in FIG. There's a problem. However, in the case shown in FIG. 13, it can be understood that the settling time can be shortened because the position deviation does not exceed the settling width once within the settling width.

<< Third Embodiment >> Next, a servo motor control device according to a third embodiment of the present invention will be described with reference to the drawings. When performing position control of control objects that move up and down against gravity like an elevator, there is a problem in reducing the integral value for these control objects as in the first and second embodiments described above. That is, when the integral value is reduced, the input that compensates for the influence of gravity disappears, so there is a problem that when the influence of gravity is greater than the frictional force, it shifts downward. Therefore, in the third embodiment,
A description will be given of a servo motor control device capable of shortening the settling time by compensating the gravitational term with a disturbance observer and allowing the position deviation to remain in the settling width even if the integral value is reduced.

FIG. 19 is a control block diagram of the servo motor control apparatus according to the third embodiment. 19, components having the same functions and configurations as those of the control devices of the first and second embodiments described above are designated by the same reference numerals, and the description thereof will be omitted. The servo driver 33 in the third embodiment is similar to the above-described first embodiment in that the position deviation calculator 5, the position deviation proportionalizer 6, the position deviation integrator 7, the speed FF calculator 8, the speed calculator 10, and the speed calculator Deviation calculator 11, speed deviation proportional unit 1
2, a speed deviation integrator 13, a current command calculator 14, and a current controller 15. In the servo motor control device according to the third embodiment, the configuration different from that of the first embodiment is as follows.
This is the point that a speed feedforward filter (hereinafter abbreviated as speed FF filter) 29, a timing generator 38, a disturbance observer 40, and a torque filter 48 in the servo driver 33 are provided. As shown in FIG. 19, the disturbance observer 40 is configured to receive the position signal θ from the position detector 2 and the current command calculator I * and output the final current amount It * to the torque filter 48. The disturbance observer 40 includes a torque calculator 41, an acceleration calculator 42, a subtractor 43, a low-pass filter 44, a current amount converter 45, and an adder 46.

The operation of the servomotor control device of the third embodiment having the above-described configuration will be described with reference to FIG. First, the operation of the disturbance observer 40 will be described. Torque calculator 4
1 is the current command value I * output from the current command calculator 14
Is multiplied by the torque constant Kt to output the calculated torque value T *. The acceleration calculator 42 calculates the acceleration by performing the backward difference twice on the position signal θ output from the position detector 2 which is an encoder. Further, the acceleration calculator 42 calculates and outputs the required torque Ti when the disturbance does not act by multiplying the inertia moment of the servo motor 1 by the calculated acceleration. Then, the subtractor 43 calculates the torque calculation value T * from the torque calculator 41 and the acceleration calculator 42.
From the required torque Ti to calculate the estimated disturbance value TL. The theory will be briefly described below. When the actual moment of inertia of the servomotor 1 is Jn and the disturbance amount is TL, the following equation (17) is established.

[0103]

[Equation 17]

The estimated disturbance value TL can be easily obtained by the calculation of the following equation (18).

[0105]

[Equation 18]

That is, by performing the calculation of the equation (18), the estimated disturbance value T such as gravity acting on the servo motor 1 is calculated.
L can be calculated. However, since the position signal needs to be differentiated twice in order to calculate the acceleration, the obtained acceleration signal becomes noisy. Therefore, the signal indicating the estimated disturbance value TL output from the subtractor 43 is the low-pass filter 44.
The high frequency component is cut by passing through. Then, the current amount converter 45 divides the filtered disturbance estimated value TL by the torque constant Kt to compensate the disturbance estimated amount I.
Output h. The adder 46 adds the current command I * before the disturbance amount compensation and the compensation current Ih to add the final current amount It.
Output * to the torque filter 48.

FIG. 20 shows a velocity command and a position deviation which are differential values of the position command when the gravity is compensated by the disturbance observer 40 with respect to the controlled object to which the gravity acts. The position deviation does not fluctuate even when the movement command is zero (point D), the position deviation shifts while being held within the settling width, and the settling time is kept at “0”. In this case, since the gravitational term is compensated by the disturbance observer 40, even if the position integral term is decreased when the movement command is zero as in the above-described embodiment, the positional deviation does not occur in the third embodiment. . FIG. 21 is shown for comparison with FIG. 20, and shows the speed command and the position deviation which are the differential values of the position command when the disturbance observer is not provided for the same control target as in FIG. .. In FIG. 21, it can be seen that the position of the servo motor 1 is returned and the position deviation θe appears to be positive because the gravity cannot be completely compensated by reducing the integral calculation value when the movement command is zero (point D).

In the first embodiment described above, the method of reducing the vibration of the position deviation by smoothing the position deviation integrated value and the speed deviation integrated value individually has been described. It goes without saying that the vibration reduction method of the first embodiment works effectively for all other embodiments. In the third embodiment, the torque filter 48 is provided between the current command calculator 14 and the current controller 15. Similarly, the torque filter 48 also prevents a sudden change in torque due to a sudden change in the position deviation integrated value, the speed deviation integrated value, or the speed FF calculation value at the final stage. That is, the timing generator 3 of the third embodiment
When the timing signal is output from No. 8, the time constant of the torque filter 48 may be set large only for a preset period. In this setting, when one or more timing signals are output, the first timing signal is used. As a result, similarly to the other embodiments, it is possible to reduce the vibration of the positional deviation and shorten the settling time for a system with small damping.

The torque filter 48 used in the third embodiment may be a moving average filter. Further, even when the torque filter 48 used in the third embodiment is used in the first and second embodiments, the input torque is also smoothed, so that the same effect as that of the third embodiment is obtained. It goes without saying that the output of the individual timing signal for each integral value or speed FF calculation value of the timing generator 37 described in the second embodiment is also effective in the third embodiment.

Next, in the configuration of the third embodiment, when the speed deviation integrator 13 is operated without operating the position deviation integrator 7, the influence of the speed FF filter 29 and the speed detection filter of the speed calculator 10 is influenced. A method of canceling the above and shortening the settling time without generating the position steady deviation during the operation will be described. The time constant of the speed FF filter 29 is Tf.
In the case of, the position deviation θe during acceleration / deceleration is obtained by the following equation (19) using the theorem of the final value.

[0111]

[Formula 19]

Similarly, when the time constant of the speed detection filter of the speed calculator 10 is Ts, the position deviation θe during acceleration / deceleration is similarly obtained by the following formula (20) using the final value theorem. .

[0113]

[Equation 20]

Equations (19) and (20) show that the positional deviation θe occurs in the opposite direction if the filter time constants are the same. The position deviation integrator 7
When is used, since the integration directly acts on the position deviation θe, the position deviation does not occur during acceleration / deceleration even when the speed FF filter processing or the speed detection filter processing is executed. Therefore, in the configuration of the third embodiment, when the speed deviation integrator 13 is operated without using the position deviation integrator 7, the time constant Tf of the speed FF filter 29 and the time constant Ts of the speed detection filter have the same value. When set to, the position deviation θe during acceleration / deceleration becomes zero. Therefore, when the speed deviation integrator 13 is operated without operating the position deviation integrator 7, the speed FF is changed when the time constant Ts of the speed detection filter is changed.
By changing the time constant Tf of the filter 9 in conjunction with each other, it is possible to suppress the occurrence of position deviation due to the influence of the filter. Here, in the case of only the speed deviation integrator 13, it is not necessary to decrease the speed FF calculation value from the speed FF filter 29 near the time when the movement command is zero, and only the speed deviation integration value needs to be decreased.

22 and 23 are waveform diagrams respectively showing the speed command and the position deviation when the position deviation integrator 7 is not provided as the integrator and only the speed deviation integrator 13 is provided. 22 shows the speed command (a) and the position deviation (b) when only the speed FF filter acts, and FIG. 23 shows the speed command (a) and the position deviation (b) when only the speed detection filter acts. ) And. 22 and 23, the time constant Tf of the speed FF filter 29 and the time constant Ts of the speed detection filter are the same value. As a result,
2 and the steady-state deviation during acceleration / deceleration in FIG. 23 have opposite signs,
It can be confirmed that the absolute values of position deviations are equal. Speed FF
By equalizing the time constants of the filter 29 and the speed detection filter, the position deviation θe during acceleration / deceleration can be set to “0” as shown in FIG.

As described above, by making the time constants of the speed FF filter 29 and the speed detection filter of the speed calculator 10 equal, the position deviation θe can be made close to zero even during acceleration / deceleration. Further, by reducing only the velocity deviation integrated value in the vicinity of the movement command zero, it becomes possible to shorten the settling time only by the velocity deviation integrator 13 when friction is small.

[0117]

As is apparent from the above detailed description of the embodiments, the present invention has the following effects. According to the present invention, a position deviation integrator and a speed deviation integrator are used together for a control object having a large viscous friction, and a timing signal from a timing generator is received in the vicinity of the movement command being “0”. By reducing the integrated value of position deviation and integrated value of speed deviation, the occurrence of position deviation during movement can be suppressed and the position deviation can be kept near zero even after the movement command is zero, and the settling time can be shortened. The device can be obtained. Further, according to the present invention, when the timing signal is output, when the position deviation occurs due to the small gain, the position deviation integrator is prevented from operating for a preset period. Thus, it is possible to obtain a servomotor control device capable of reducing the occurrence of overshoot due to the accumulation of position deviation integrated values while setting the position deviation to “0” and shortening the settling time.

Further, according to the present invention, when the speed FF filter operates, at least one of the position deviation integrated value and the speed FF calculation value from the speed FF filter is input to the timing signal from the timing generator. By reducing the time, the influence of the filter delay of the speed FF calculation value can be removed, and the settling time can be shortened without overshoot of the position deviation after the movement command is zero. . Further, according to the present invention, when the calculation delay of the position or velocity loop is large or the torque filter is large, the timing of decreasing the position deviation integral term is made earlier than the time when the movement command is zero, thereby causing the deviation of the position deviation. Can be reduced to shorten the settling time. Further, according to the present invention, when the rigidity is low and the gain cannot be set high, the output of the timing signal for decreasing the position deviation integrated value output from the position deviation integrator and the speed FF calculation value output from the speed FF filter is output. By changing the time from when the movement command is zero to before and after it, the position deviation that occurs when the movement command is zero can be canceled, the position deviation that occurs when the movement command is zero can be reduced, and the settling time can be shortened. it can.

Further, according to the present invention, the condition for outputting the timing signal for reducing the value of the position deviation integrator or the speed FF filter according to the position deviation amount and the deviation direction in the vicinity where the movement command becomes zero. By providing the constant setting means for determining the constant of the equation in the timing generator, the position deviation can be automatically reduced even after the movement command is zero. Further, according to the present invention, by making the time constants of the speed FF filter and the speed detection filter equal, the position deviation can be made close to zero even during acceleration / deceleration, and only the speed deviation integrated value is zero when the movement command is zero. When the friction and the like are small, the settling time can be shortened only by the velocity deviation integrator by decreasing the value in the vicinity of.

Further, according to the present invention, when the timing signal is output from the timing generator, the time constant of the torque filter is set to be large for a preset period, so that the position deviation integrated value or the speed deviation is increased. It is possible to prevent a sudden change in the torque due to a sudden change in the integrated value or the speed FF calculation value, reduce the vibration of the position deviation and reduce the settling time for a control system with small damping. Further, according to the present invention, the disturbance observer is added to the controlled object on which gravity acts, so that the disturbance observer is reduced even if the position deviation integrated value or the velocity deviation integrated value is reduced in the vicinity where the movement command becomes zero. Gravity is compensated by, and the settling time is greatly shortened by shifting while maintaining the settling width without swinging the position deviation.

[Brief description of drawings]

FIG. 1 is a control block diagram of a servo motor control device according to a first embodiment of the present invention.

FIG. 2 is a flowchart of a position / speed loop according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a differential value and a position deviation of a position command with respect to a control object having a small viscous friction in the control device according to the first embodiment.

FIG. 4 is a diagram showing a differential value and a position deviation of a position command of a control device that does not reduce a position deviation integral value with respect to a control object having a small viscous friction.

FIG. 5 is a diagram showing a differential value and a position deviation of a position command in the control device according to the first embodiment for a control object having a large viscous friction.

FIG. 6 is a diagram showing a differential value and a position deviation of a position command of a control device having only a velocity deviation integrator and not a position deviation integrator for a control object having a large viscous friction.

FIG. 7 is a control block diagram of a servo motor control device according to a second embodiment of the present invention.

FIG. 8 is a flowchart of a position / speed loop according to the second embodiment.

FIG. 9 is a diagram showing a differential value and a position deviation of a position command of the control device that does not reduce the value of the speed FF filter when the movement command is zero.

FIG. 10 is a diagram showing a velocity FF calculation value and a position deviation integrated value in the vicinity where the movement command becomes zero.

FIG. 11 is a flowchart of the position / speed loop of the present invention.

FIG. 12 is a diagram showing a differential value and a position deviation of a position command of the control device according to the second embodiment of the present invention when the gain is small.

13 is an enlarged view of the positional deviation of FIG.

FIG. 14 is a diagram showing a position deviation integral value and a velocity FF in the vicinity of a zero movement command of the control device of the present invention when the gain is small.
It is a figure which shows a calculated value.

FIG. 15 is a diagram showing a differential value and a position deviation of the position command when the position deviation integrated value and the speed FF calculation value are simultaneously decreased when the movement command is zero when the gain is small.

16 is an enlarged view of the position deviation of FIG.

FIG. 17 is an enlarged view of the position deviation of the control device that reduces the position deviation integral value only once when the gain is small.

FIG. 18 is a diagram showing a position deviation integrated value and a speed FF calculation value of a control device that reduces the position deviation integrated value only once when the gain is small.

FIG. 19 is a control block diagram of a servo motor control device according to a third embodiment of the present invention.

FIG. 20 is a diagram showing a differential value and a position deviation of a position command of the control device according to the third embodiment with respect to a controlled object on which gravity acts.

FIG. 21 is a diagram showing a differential value and a position deviation of a position command of a control device without a disturbance observer with respect to a controlled object on which gravity acts.

FIG. 22 is a diagram showing a position deviation when a speed FF filter acts on a control device without a position deviation integrator.

FIG. 23 is a diagram showing a position deviation when a speed detection filter acts on a control device without a position deviation integrator.

FIG. 24 is a block diagram of a conventional servo motor control device.

FIG. 25 is a flowchart of a conventional position / speed loop.

[Explanation of symbols]

2 Position detector 3,23,33 Servo driver 5 Position deviation calculator 6 Position deviation proportional device 7 Position deviation integrator 8 Speed feedforward calculator 9 Velocity feedforward filter 11 Speed deviation calculator 12 Speed deviation proportional device 13 Speed deviation integrator 14 Current command calculator 17, 37, 38 Timing generator 40 Disturbance Observer 41 Torque calculator 42 Acceleration calculator 43 Subtractor 44 low pass filter 45 Current converter 46 adder

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Toru Tazawa             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. (72) Inventor Tomokuni Iijima             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. (72) Inventor Kenichi Suzuki             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. F term (reference) 5H303 CC05 CC07 DD01 KK03 KK11                       KK22 KK24 KK28                 5H550 BB10 FF05 GG01 GG03 JJ03                       JJ23 JJ24 JJ26 LL08 LL35

Claims (16)

[Claims] 1. A position command generator, a timing generator for generating a timing signal in the vicinity of a movement command output from the position command generator, which is a change amount of the position command, becomes zero, and the movement command and the servo. A position deviation integrator that performs an integral calculation of a position deviation that is a difference from the position of the motor, outputs a position deviation integrated value, and decreases the position deviation integrated value when a timing signal from the timing generator is input, The speed deviation obtained by the movement command and the speed of the servo motor, which is the difference between the speed of the servo motor and the speed deviation, are integrated to output the speed deviation integrated value, and the timing signal from the timing generator is input. And a speed deviation integrator that reduces the speed deviation integrated value. 2. The servo motor control device according to claim 1, wherein the speed deviation integrator is arranged in a speed loop formed inside the position loop having the position deviation integrator. 3. At least one of a position deviation integration calculation and a speed deviation integration calculation for at least one of a position deviation integrator and a speed deviation integrator for a certain period after a timing signal from a timing generator is input. 2. The servomotor control device according to claim 1, wherein the servomotor is stopped. 4. The position deviation integrator gradually reduces the position deviation integrated value by using a filter means when the timing signal from the timing generator is input, and the speed deviation integrator causes the timing deviation signal from the timing generator. 2. The servo motor control apparatus according to claim 1, wherein the integrated value of the speed deviation is gradually reduced by using a filter means when is input. 5. A position command generator, a timing generator that generates at least one timing signal in the vicinity of a movement command output from the position command generator, which is a variation of the position command, becoming zero, and the movement. Position where the integral calculation of the position deviation, which is the difference between the command and the position of the servo motor, is performed to output the position deviation integrated value, and when the timing signal from the timing generator is input, the position deviation integrated value is decreased. A deviation integrator; and a speed feedforward device that calculates a speed feedforward value based on the movement command and decreases the speed feedforward value when a timing signal from the timing generator is input. Characteristic servo motor control device. 6. A position command generator, a timing generator that generates a timing signal before a movement command, which is a change in the position command output from the position command generator, becomes zero, the movement command and the servo. When the timing signal from the timing generator is input before the movement command becomes zero and the position deviation integral value is output by performing the integral calculation of the position deviation which is the difference from the motor position, the position deviation integration is performed. A servomotor control device comprising: a position deviation integrator that decreases the value; and a speed feedforward device that calculates a speed feedforward value based on the movement command and outputs the feedforward value. 7. When the timing generator individually outputs a timing signal for reducing the velocity feedforward value of the velocity feedforward device and the position deviation integral value of the position deviation integrator, when the movement command is near zero, the position When the position deviation, which is the difference between the command and the actual position of the servomotor, is positive, one timing signal that decreases the position deviation integral value is output near the movement command becoming zero, and the speed feedforward value is output. The other timing signal that decreases the output is delayed relative to the one timing signal, or if the position deviation is negative near the movement command becomes zero,
One of the timing signals for decreasing the speed feedforward value is output in the vicinity of the movement command becoming zero,
7. The servo motor control device according to claim 5, wherein the other timing signal for reducing the position deviation integrated value is output after being delayed with respect to the one timing signal. 8. When the timing generator individually outputs a timing signal for reducing the velocity feedforward value of the velocity feedforward device and the position deviation integral value of the position deviation integrator, a position command near the movement command becomes zero. A constant setting means for calculating and setting a judgment constant for carrying out the judgment to output the individual timing signals based on a position deviation which is a difference between the actual position of the servo motor and the actual position of the servo motor. Item 7. A control device for a servo motor according to item 5 or 6. 9. The position deviation integrator gradually reduces the position deviation integrated value by using a filter means when the timing signal from the timing generator is input, and the speed feed forward device causes the speed feed forward device to decrease the timing signal from the timing generator. 7. The servo motor control device according to claim 5, wherein the speed feedforward value is gradually reduced by using a filter means when is input. 10. The position deviation integrator stops the integration calculation of the position deviation of the position deviation integrator for a certain period after the timing signal from the timing generator is input. Or a servo motor control device according to any one of 6). 11. A position command generator, a timing generator for generating a timing signal in the vicinity of which a movement command output from the position command generator, which is a change amount of the position command, becomes zero. The integrated speed deviation, which is the difference between the speed command and the servo motor speed, is integrated and the integrated speed deviation value is output. When the timing signal from the timing generator is input, the integrated speed deviation value is calculated. A speed deviation integrator that reduces the speed, a speed feed forwarder that calculates a speed feed forward value by performing a filtering process of a preset time constant based on the movement command, and a servo motor speed of the speed feed forwarder. A speed detector that performs a filtering process using a filter time constant that is similar to the filter time constant and outputs a speed calculation value. Control device for a servo motor, characterized in that. 12. A torque filter for increasing a time constant of a filter for smoothing a torque command when a timing signal output from a timing generator is obtained, the torque filter being provided for a certain period. The control device for the servomotor according to any one of 6 and 11. 13. The servo motor control device according to claim 1, further comprising a disturbance observer for estimating and canceling a disturbance acting on the servo motor. 14. A first step of generating a timing signal in the vicinity of a movement command, which is a variation of the position command, becoming zero, and when the timing signal is input, the movement command and the position of the servomotor A second step of performing an integral calculation of the position deviation which is a difference to reduce the position deviation integrated value; and, when the timing signal is input, the speed command obtained from the movement command and the servo motor And a third step of reducing the integrated value of the speed deviation by performing an integral calculation of a speed deviation which is a difference with the speed. 15. At least one of a second step and a third step stops at least one of a position deviation integral calculation and a speed deviation integral calculation for a certain period after the timing signal is input. 15. The method of controlling a servomotor according to claim 14, wherein: 16. The position deviation integral value is gradually decreased when the timing signal in the first step is input,
15. The servo motor control method according to claim 14, wherein the integrated value of the speed deviation is gradually reduced when the timing signal is input.