JP2709766B2 - Feed forward control method of servo motor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control system of a servomotor for driving a feed shaft of a machine tool and a robot arm, and more particularly to a feedforward control of the servomotor.

[0002]

2. Description of the Related Art When a servomotor is used to control a feed axis of a machine tool, a robot arm, or the like, feedforward control is performed to reduce the amount of positional deviation.
In particular, when high-speed cutting is performed by a machine tool, a shape error occurs due to a delay in following the servo system. Therefore, in order to reduce this shape error, feedforward may be applied to the position loop. That is, for each position and speed processing cycle, the movement command is differentiated, and a value obtained by multiplying the differentiated value by a feedforward coefficient is added to the speed command obtained by ordinary position loop processing. , The position deviation is reduced, and the servo delay is corrected.

However, the distribution cycle (ITP cycle) in which the movement command is transferred from the numerical controller or the like to the servo circuit side, that is, the position loop, is about 8 msec, and the cycle of the position loop and the speed loop inside the servo circuit is 2msec
Alternatively, it is 1 msec. In the position loop, the ITP cycle in which the movement command is transferred from the numerical controller is divided by the position loop cycle, and control is performed so that the movement commands in each of the divided position loop cycles are equal. Therefore, even if the acceleration / deceleration time constant is given to the movement command output from the numerical controller and the movement command is output, in the position loop, the movement command in each position loop processing cycle TP for each position loop processing cycle TP in each ITP cycle. Is controlled so as to be equal, a large step is generated in the movement command only between the movement commands of the position loop processing in which the ITP cycle changes, and this is differentiated by the feedforward term to become a large value, and the speed command changes the high frequency component. This means that it is not possible to follow in the speed loop, and the position deviation swells, causing a large shock to the movement of the motor or the machine. Therefore, in order to eliminate this drawback, a servo motor control method for performing a smoothing operation (smoothing process) to eliminate the swell by inserting acceleration / deceleration processing in the feedforward term of position control and further speed control. The inventor of the present invention has proposed in Japanese Patent Application No. 1-150481.

[0004]

However, the above-mentioned smoothing operation is for averaging the past data in time, and in order to cause a time delay equivalently in the feedforward term, the position loop control is performed. In the system, swelling of the position deviation occurs at the time of acceleration / deceleration.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above-mentioned problems and to provide a servomotor control method for improving the command followability of a servo system.

[0006]

According to the present invention, when the number of divisions obtained by dividing the ITP cycle by the position / velocity loop processing cycle is represented by N, N
The average value of the movement command in each position / speed loop processing cycle is calculated, and the feedforward amount of the position obtained by multiplying this average value by the position feedforward coefficient is added to the speed command obtained by the position loop processing to obtain the speed. This is a speed command for loop processing. Further, a differential value of the average value in the position / velocity loop processing advanced by a set number of cycles from the position / velocity loop period is obtained, and a velocity feedforward obtained by multiplying the differential value by a velocity feedforward coefficient is obtained. The above problem was solved by adding the amount to the torque command value obtained by the speed loop processing to obtain a torque command to the servomotor.

[0007]

The movement command of the ITP cycle one cycle ahead of the ITP cycle is read, and the movement command of the position / velocity loop processing cycle is obtained. N is the number obtained by dividing the P cycle by the position / speed loop processing cycle, α1 is the feedforward coefficient of the position, P is the coefficient that converts the unit of the pulse of the movement command into the unit of the speed command, and the position / speed loop processing cycle Assuming that the movement command at j is a (j), the feedforward amount FF at the position in the j period is obtained from the average value b (j) of the movement commands in N periods around the j period. However, since the ITP cycle is usually 8 msec and the position / velocity loop cycle is usually 1 msec or 2 msec, the division number N is an even number, so that (−N / 2) +1 to N / sec
2 or average value b (j) of -N / 2 to (N / 2) -1
Ask by. For example, the average value of the period from (−N / 2) to N / 2-1 is the following equation 1.

[0008]

## EQU1 ## The average value of the period of (-N / 2) +1 to (N / 2) is given by the following equation (2).

[0009]

## EQU2 ## Then, the average value b (j) is multiplied by the coefficients P and α1, and the feedforward amount FF at the position is obtained by the following equation (3).

[0010]

[Mathematical formula-see original document] Further, a weighted average of the values obtained by the calculations of the above formulas 1 and 2 is obtained to obtain an average value b (j), which is multiplied by a coefficient P and α1 to obtain a feedforward amount FF of the position. Is also good. The operation of the above equation 1 or 2 and further the equation 1
The feed-forward amount FF at the position obtained by the weighted average of the values obtained by the calculation of Expression 2 is added to the speed command obtained by the position loop processing of the cycle j, and the added value is added to the speed command for the speed loop processing. I do. Further, a coefficient for converting the unit of the pulse of the movement command into the unit of the speed command into a differential value of the average value b (j + L) obtained by the above equation 1 in the cycle advanced by L cycles is P, the speed feedforward coefficient α2 and Coefficient P for converting pulse unit of movement command to current unit
'To multiply the speed feedforward amount FFv by the following equation 4.
Ask by

[0011]

[Mathematical formula-see original document] The speed feedforward amount FFv determined by the above expression (4) is added to the torque command (current command) determined by the speed loop processing to obtain a torque command to the servomotor.

[0012]

FIG. 1 is a block diagram of a servo system according to an embodiment of the present invention. In FIG. 1, KP of the transfer function 1 is a position gain in the position loop, transfer function 2 is an integral term of the speed loop, Ts is a period of the position / speed loop processing, and k1 is an integral gain in the speed loop. K of transfer function 3
2 is a proportional gain in the speed loop, 4 is a mechanical part of the servo motor, Kt is a torque constant, Jm is inertia, 5
Is a transfer function for calculating the position by integrating the rotation speed of the servomotor. Reference numerals 6 and 7 denote position feedforward terms, 6 denotes a smoothing circuit, and 7 denotes a term multiplied by the position feedforward coefficient α1.

Reference numeral 8 denotes a transfer function for delaying the output of the smoothing circuit 6 by a set number of position / velocity loop processing periods, reference numeral 9 denotes a transfer function for performing differentiation, and reference numeral 10 denotes a term multiplied by a speed feedforward coefficient α2. This element 8, 9,
At 10, speed feed forward control is performed.
Numeral 11 denotes a DDA (Digital Different) for dividing a movement command MCMD sent from a CNC (numerical control device built in a computer) for each distribution cycle (ITP cycle) into a movement command for each position / speed loop processing cycle.
ial Analyzer).

The transfer functions 2, 8,
Reference numeral 9 denotes a discrete value control system. C, which is the host controller
N outputs a movement command MCMD for each ITP cycle, and the DDA 11 obtains a movement command a in the position / speed loop cycle. Further, in order to obtain the feedforward amount FF from the above elements 6 and 8, it is necessary to obtain a movement command for the next ITP cycle, and the DDA 11 obtains a movement command for the next ITP cycle. . Then, a movement command a in the position / speed loop processing is obtained, and a feedback amount Pf of an actual movement amount of the servo motor is subtracted from the movement command a in the position / speed loop cycle to obtain a position deviation. The value multiplied by Kp is subjected to the operation of Expression 1 or 2 by the elements 6 and 7, or the position feedforward amount FF obtained by the weighted average of the values obtained by the operation of Expression 1 or Expression 2 is added. A speed command is obtained, a feedback amount Vf of the actual speed of the servo motor is subtracted from the speed command, and a speed deviation is obtained. Is multiplied by a proportional constant k2, and the same integral and proportional speed loop processing as in the prior art are performed to obtain a torque command. Then, the speed feedforward amount FFv is obtained by performing the calculation of Expression 4, and the speed feedforward amount FFv is added to the torque command obtained by the same processing as that of the related art to obtain a feedforward controlled torque command. Drive the motor.

FIG. 2 is a block diagram of a digital servo control device for implementing one embodiment of the present invention. The configuration is the same as that of a conventional device for performing digital servo control.
It is shown schematically. In FIG. 2, reference numeral 20 denotes a numerical controller (NC), 21 denotes a shared RAM, and 22 denotes a processor (C).
PU), a digital servo circuit having ROM, RAM, etc., 23 is a servo amplifier such as a transistor inverter,
Reference numeral 24 denotes a servo motor, and reference numeral 25 denotes a pulse coder that generates a pulse with the rotation of the servo motor 24.

The NC 20 writes the movement command MCMD into the shared RAM 21 at every IPT cycle (distribution cycle), and the CPU of the digital servo circuit 22 writes the movement command MCMD into the shared RAM 21.
The position / velocity loop processing is performed at a period Ts (ITP = Ts × N) obtained by reading from the AM 21 and dividing the ITP period into N pieces. For each ITP cycle, a movement command a (j) in the position / velocity loop processing Ts is determined so that the movement command MCMD output from the NC 20 is evenly distributed during the ITP cycle, and the movement command a (j) and the pulse coder 25 Servo motor 24 obtained by the feedback pulse of
Of the servo motor 24 obtained by the speed command and the feedback pulse from the pulse coder 25 while performing a position loop process based on the difference between the current position and the current position. A speed loop process and a speed feedforward process are performed from the actual speed to obtain a torque command (current command). Then, a current loop process and further a current feedforward process are performed, a PWM command is created, and the servo motor 24 is driven via the servo amplifier 23.

FIGS. 3 and 4 are flowcharts of the movement command reading process for each ITP cycle, the position loop processing and the speed loop processing for each position / speed loop processing cycle, which are executed by the CPU of the digital servo circuit 22. In this embodiment, the position / velocity loop processing is performed by dividing the IPT cycle into four (N = 4). Further, in order to obtain the position feedforward amount FF, it is assumed that the average value b (j) of the movement commands of N cycles around the cycle is obtained by using the above equation 1, and the speed feedforward processing An example in which two cycles (L = 2) are advanced will be described below. Therefore, the average value b (j) of the movement command is given by the following equation 5, and the feedforward amount FF is given by equation 6.

[0018]

(Equation 5)

[0019]

## EQU6 ## The feedforward amount FFv of the speed is given by the following equation (7) from the equation (4).

[0020]

The position / velocity loop processing executed by the CPU of the digital servo circuit 32 will be described with reference to the flowcharts of FIGS. First, the CPU reads the next movement command MCMD output from the NC 20 every ITP cycle via the shared RAM 21 (step S1). In addition, actually,
Since the position / velocity loop processing is performed by the movement command of the position / velocity loop processing cycle delayed by one cycle in the velocity loop processing,
Movement command MCM output from NC20 every ITP cycle
D will be read.

Then, for each position / velocity loop processing cycle, C
The PU starts the processing shown in FIG. 4, and calculates a movement command a (j + 4) for each position / speed loop from the distribution movement command MCMD read for each ITP cycle (step S10).
Then, the storage contents of the registers R (j−2) to R (j + 4) are shifted by one, and the movement command a (j + 4) calculated in step S10 is stored in the rayster R (j + 4). At the time when the distribution movement command MCMD is not output,
The content stored in each register is "0". That is, the contents stored in the register R (j-1) are stored in the register R (j-2).
Similarly, the stored contents of R (j) are stored in R (j-1), the stored contents of R (j + 1) are stored in R (j), and R (j + 1) is stored.
Is stored contents of R (j + 2), and R (j + 2) is stored in R (j +
3), the storage content of R (j + 4) is stored in R (j + 3), and the movement command a (j) is stored in the register R (j + 4).
+4) (steps S11-1 to S11-).
7). Then, the movement command a (j) of the cycle stored in the register R (j) is added to the position deviation e (j-1) of the previous position / speed loop cycle stored in the register, and the servo detected by the pulse coder 25 is added. The position deviation of the cycle is obtained by subtracting the feedback amount Pf (j) of the position of the movement amount of the motor (step S12).

Since the distribution movement command MCMD is to be read one ITP cycle earlier, the movement command a (j) of the position / speed loop cycle is the movement command a (j + 4) calculated in the current cycle. R (j + 4) that stores
It is stored in the register R (j) four places earlier. Next, the registers R (j + 1), R (j), R (j-1), R
Movement command a (j + 1), a stored in (j-2)
(J), a (j-1), and a (j-2) are used to calculate the average value b (j), and calculate the average value b (j).
To obtain the position feedforward amount FF (step S13). Next, the position feedforward amount FF obtained at step S13 is multiplied by the value obtained by multiplying the position deviation e (j) obtained at step S12 by the position loop gain Kp. The added speed command Vc (j) is obtained (step S14), and the speed command Vc (j) is used to perform the same speed loop processing as before to obtain the torque command Tc '(j) before correction by speed feedforward (step S14). S15). Next, the speed feedforward amount FF is calculated by performing the calculation of Expression 7 using the movement commands a (j + 3) and a (j-1) stored in the registers R (j + 3) and R (j-1).
v is obtained (step S16). Then, step S1
5 is added to the pre-correction torque command Tc ′ (j) calculated in step S15 to obtain the torque command Tc (j) by adding the speed feedforward amount FFv calculated in step S15 (step S1).
7), the torque command Tc (j) is transferred to the current loop processing (step S18), and the processing of the position / velocity loop ends.

Tables 1, 5 and 6 show an example of the present embodiment. As shown in FIG. 5A, a movement command MCMD (n) for each ITP cycle is input, and A movement command a (j) is obtained for each speed loop processing cycle,
When an average value b (j) (a position feedforward amount is obtained by multiplying the average value by the coefficients α1 and P) from the movement command a (j) is obtained as shown in FIG.
Table 1 shows that each of the registers R (j + 4) to R (j
2 (b) shows the values of the values a (j + 4) to a (j-2) and the average value b (j) stored in -2). For reference, the average value b (j) obtained by Expression 2 is calculated as b (j) ′ Expression 1,
The weighted average value of the average calculated by Equation 2 is shown as b (j) ″. In Table 1, +4 indicates the register R (j +
4) The value of a (j + 4) stored in 4), +3 is the register R
The value of a (j + 3) stored in (j + 3),... J is the value of a (j) stored in register R (j),. ).

[0024]

[Table 1] As can be seen from FIG. 5 and Table 1, even if the movement command a (j) changes abruptly every ITP cycle, the position feedforward amount FF (average value b (j)) does not change abruptly. This means that there is no abrupt change as in the case where the feed command is multiplied by a coefficient obtained by differentiating the movement command as in the conventional feed forward control, and therefore the undulation of the position deviation is also eliminated. FIG. 6 is a graph showing a position deviation when a ramp input is performed on a movement command in a conventional feedforward control in which a coefficient obtained by differentiating a movement command is multiplied by a coefficient. In the present invention, when data is not advanced by the speed feedforward control (L = 0), the movement command is a graph obtained by measuring the position deviation at the ramp input, and FIG. 8 shows the data in the present invention. 2 cycles (L
= 2) It is a graph of the measurement result of the positional deviation when the operation is advanced. As is apparent from FIGS. 6 to 8, it is understood that the undulation of the position deviation is improved in the feedforward control of the present invention as compared with the conventional feedforward control.

In the above embodiment, the cycle j + 1 which is one cycle (future) ahead of the cycle j of the position / velocity loop, the cycle j, one cycle and j-1 and j-j which is two cycles before (the past). 2
Although the embodiment in which the movement commands are averaged has been described, a method of averaging the movement commands in the periods j + 2, j + 1, j, j-1 according to Equation 2 may be used. The average value in this case is b (j) in Table 1.
5 (b), and is shifted to the left by one cycle Ts in FIG.

As described above, when the number N obtained by dividing the ITP cycle by the position / velocity loop cycle is an even number, the position / velocity loop cycle serving as the center of averaging is completely N position / velocity loop cycles. Not the center. Therefore, the position feedforward amount FF is delayed or advanced by a half cycle of the position / velocity loop cycle. This point is evident from Table 1 and FIG. 5.
As shown in (a), the movement command MCMD of the ITP cycle is 1
It is assumed that a 4 × 4 movement command is issued only for the period. At this time, assuming that the feedforward amount is obtained by calculating the average value by Expression 1, if the position / speed loop cycle in the ITP cycle is j = 0 and the movement command a in each cycle is obtained, the average value is as shown in FIG. As shown in FIG.
1, 2, 3, 4, 4, 3, 2, 1 up to = 5. On the other hand, when the average value is obtained by Expression 2, as shown in FIG. 9C, from the period j = -2 to j = 4, 1, 2, 3, 4, 3, 2,
In FIG. 9B, it is delayed by a half cycle, and in FIG. 9C, it is advanced by a half cycle. Then, when a weighted average of the average values obtained by Expression 1 and Expression 2 is taken, as shown in FIG. 9D, (1/2), (3 /
2), (5/2), (7/2), (7/2), (5 /
2), (3/2), (1/2), and the time delay and advance are equivalently eliminated.

A process for obtaining and controlling the position feedforward amount FF based on the weighted average will be described with reference to FIGS. This embodiment is different from the above-described embodiment in that the average value is not obtained from the movement command a of the position / speed loop cycle obtained by the DDA 11, but is obtained from the movement command MCMD issued every ITP cycle.

The CPU of the digital servo circuit 22 executes the processing of FIG. 10 for each distribution cycle. First, the value stored in the register R2 is stored in the register R3, and the value of the register R is stored in the register R2.
1 is stored in the register R1, and the shared RAM is stored in the register R1.
The first move command MCMD read from 21 is stored (steps T1 and T2). Then, set the counter C to “0”
(Step T3). That is, the register R
3 is a movement command M one time before (past) in the ITP cycle.
The CMD is stored, the movement command MCDM of the cycle is stored in the register R2, and one time ahead (future) is stored in the register R1.
Is stored. In practice, the value stored in the register R2 is simply stored in the register R3,
The value stored in the register R1 is stored in the register R2,
What is necessary is just to store the distributed movement command MCDM in the register R1, and the processing is executed with a delay of one ITP cycle in the position / speed loop processing as described later. Note that the registers R1 to R3 initially store "0" in the initial setting.

On the other hand, after the CPU of the digital servo circuit 22 executes the position deviation calculation processing (the processing of step S12 in FIG. 4) for each position / speed loop processing cycle,
The processing shown in FIG. 11 is executed (only the processing for obtaining the position feedforward amount is shown in FIG. 11 and other processing is omitted). It is determined whether the value is smaller than 1/2 of the value N (= distribution period / position / velocity loop processing period) divided by the processing period (step T10). If smaller, the accumulator SUN stores the value of the register R2 from the value of the register R2. Is added (step T11). If the value of the counter C is equal to or more than N / 2, a value obtained by subtracting the value of the register R2 from the value of the register R1 is added to the accumulator SUN (step T12). Move to step S13. The accumulator SUN is initially set to "0" in the initial setting.

In step T13, the accumulator S
The value of UN is divided by the square of the number of divisions N and the average value b (j)
Ask for. Then, “1” is added to the counter C, and the average value one cycle before and stored in the register R (b) is compared with the average value in step T1.
The average value b (j) obtained in step 3 is added and divided by "2" to obtain a weighted average b (j) "(step T15). Next, the average value b (j) obtained in step T13 is calculated. It is stored in the register R (b) (step T16), and step T15
The weighted average b (j) "obtained by the above is multiplied by α1 for the feedforward coefficient of the position and P for the coefficient for converting the unit of the pulse of the movement command into the unit of the speed command to obtain the feedforward amount FF (step T17). , The following step S in FIG.
14 and below are executed. Note that, in this case, the speed feedforward amount FFv obtained in step S16 is obtained by storing the weighted average b (j) ″ obtained in step S15 and performing the calculation of Expression 4.

In the above processing, steps ST10-T
Reference numeral 13 denotes a process for calculating the average value b (j). It will be described below that this process becomes the average value of the movement commands a of N position / speed loop periods around the position / speed loop period. I do. For example, as shown in FIG. 12, the movement command MCND for each ITP cycle is A0, A1, A2, A3, A4.
Are output and A0 = A1 = 0, and A2 and A3 are
Movement commands divided for each speed loop cycle are a20, a21, a
Assuming that 22, a23, a30, a31, a32 and a33 (the number of divisions N is "4"), the average value is as follows.

N MCMD C SUM b (j) 1 A1 0 0 + A0-A1 = 0 0 2 0 + A2-A1 = A2 A2 / 16 = a20 / 4 3 A2 + A2-A1 = 2A2 2A2 / 16 = ( a20 + a21) / 4 2 A2 0 2A2 + A2-A1 = 3A2 3A2 / 16 = (a20 + a21 + a22) / 4 13A2 + A2-A1 = 4A2 4A2 / 16 = (a20 + a21 + a22 + a23) / 4 2 4A2 + A3-A2 = 3A2 + A3 (3A2 + A3) / 16 = (a21 + a22 + a23 + a30) / 4 3 3A2 + A3 + A3-A2 = 2A2 + 2A3 (2A2 + 2A3) / 16 = (a22 + a23 + a30 + a31) / 4 3 A3 0 2A2 + 2A3 + A3-A2 = A2 + 3A3 (A2 + 3A3) / 16 = (a23 + a30 + a31 + a32) / 4 1 A2 + 3A3 + A3-A2 = 4A3 4A3 / 16 = (a30 + a31 + a32 + a33) / 4... As described above, the average value b (j) can also be obtained by the processing in steps ST10 to T13. Then, if a weighted average of the average value b (j) obtained in the position / speed loop cycle and the average value obtained one cycle before the position / speed loop cycle is obtained, the feedforward amount FF without delay and advance is obtained. be able to. In the above example, for example, the average value b (j) obtained in the first position / speed loop cycle (counter C = 0) in the ITP cycle n = 3 is the movement command a30 and 1 for the position / speed loop cycle. Movement command a23 in the previous cycle and move command a in the next and next cycles
31 and a32 are added. That is, the calculation of Expression 2 is performed. On the other hand, the average value b (j) obtained in the previous position / velocity loop cycle is (when n = 2, C = 3)
The movement command a30 of the speed loop cycle, the movement commands a23 and a22 of one and two previous cycles, and the movement command a of the next cycle
31 is added, and the average value is obtained by the operation of Expression 1. Therefore, the weighted average b (j) ″ may be obtained by adding the average calculated in the cycle and the average calculated in the previous cycle and dividing by 2.

In the above-described embodiment, the case where the ITP cycle is divided into four parts has been described. However, even in cases other than the case where the ITP cycle is divided into four parts, the movement command of each period of N divisions before and after the position / velocity loop cycle is used. a may be averaged.

[0034]

According to the present invention, a feedforward control is performed by obtaining a feedforward amount by also taking in a move command issued after the move command in the position / velocity loop processing. Even if the distribution movement command from the host controller fluctuates stepwise, the undulation of the position deviation is reduced, and the shock applied to the movement of the motor and the machine is reduced. If the number obtained by dividing the distribution cycle by the cycle of the position / velocity loop processing is an even number, when averaging the movement commands of the number of preceding and succeeding divisions by adding the cycle, the average of the movement commands before the cycle is taken. By obtaining a weighted average of the average values when one cycle is taken more (after the future) and one cycle after the (past), it is possible to obtain a position feedforward amount with a time delay and no advance equivalently. And more precise control.

[Brief description of the drawings]

FIG. 1 is a block diagram of a servo system according to an embodiment of the present invention.

FIG. 2 is a block diagram of a digital servo control device implementing the embodiment.

FIG. 3 is a flowchart of a process executed by a processor of a digital servo circuit for each ITP cycle in the embodiment.

FIG. 4 is a flowchart of a position / velocity loop process in the embodiment.

FIG. 5 is an explanatory diagram of an example of a position feedforward amount (average value) in the embodiment.

FIG. 6 is a diagram showing a transition of a position deviation when a ramp is input to a movement command in a conventional servo system that performs feedforward control.

FIG. 7 is a view showing a transition of a position deviation when a ramp amount is input to a movement command with the advance amount L of the speed feedforward control set to 0 in one embodiment of the present invention.

FIG. 8 is a view showing a transition of a position deviation when a ramp amount is input to a movement command with an advance amount L of speed feedforward control set to 2 in one embodiment of the present invention.

FIG. 9 is an explanatory diagram of an average value by a smoothing process.

FIG. 10 is a flowchart of a process executed by a processor of a digital servo circuit for each ITP cycle according to another embodiment of the present invention.

FIG. 11 is a flowchart of a process of obtaining a position feedforward during a position / velocity loop process in another embodiment.

FIG. 12 is an explanatory diagram of a process for obtaining an average value.

[Explanation of symbols]

1 Term of transfer function of position loop processing 2 Term of transfer function of integral of speed loop processing 3 Term of transfer function proportional to speed loop 4 Term of transfer function of motor 5 Term of transfer function converting from speed to position 6 Position Feedforward smoothing circuit 7 Position feedforward position feedforward coefficient term 8 Speed feedforward lead element term 9 Speed feedforward derivative term 10 Speed feedforward speed feedforward coefficient term 11 DDA (Digital Differential Analyze) ) 20 Numerical controller 21 Shared RAM 22 Digital servo circuit 23 Servo amplifier 24 Servo motor 25 Pulse coder

[Formula 1]

[Formula 2]

[Equation 3]

(Equation 4)

(Equation 5)

(Equation 6)

[Formula 7]

Claims (4)

(57) [Claims] 1. A servo motor control method for obtaining a movement command in a position / velocity loop process performed for each N-divided cycle of a movement command distribution cycle from a movement command of the distribution cycle and performing position / speed loop control. The movement command of the distribution cycle one ahead of the distribution cycle is read, and the movement command in each position / speed loop processing is calculated, and the N position / velocities are centered on the cycle in the position / speed loop processing. The average value of the movement command in the loop processing cycle is obtained, the feedforward amount of the position obtained by multiplying this average value by the position feedforward coefficient is added to the speed command obtained by the position loop processing, and the speed command for the speed loop processing is added. A feed forward control method for a servo motor characterized by the following. 2. A control method of a servomotor for obtaining a movement command in a position / velocity loop process performed for each N-divided cycle of a distribution cycle of the movement command from the movement command of the distribution cycle, and performing position / speed loop control. In addition to reading and storing the movement command of the distribution cycle immediately before the distribution cycle, the movement command of the distribution cycle one past before the distribution cycle is also stored, and the movement command of the distribution cycle is stored. Based on the movement command of the distribution cycle one ahead of the distribution cycle and the movement command of the distribution cycle one before, the position and the position
The average value of the movement commands in each of the N position / speed loop processing cycles is obtained centering on the cycle in the speed loop processing, and the feedforward amount of the position obtained by multiplying this average value by the position feedforward coefficient is calculated by the position loop processing. A feedforward control method for a servomotor, wherein a speed command for a speed loop process is added to an obtained speed command. 3. When the number of divisions N is an even number, the average value is a period in the position / velocity loop processing and each of (N / 2) -1 past and N / 2 prior to the period. position·
The average value of a total of N movement commands in the speed loop processing cycle, the cycle in the position / speed loop processing, and N / 2 and (N / 2) -1 previous positions / speeds past the cycle. 3. The feedforward control method for a servo motor according to claim 1, wherein a weighted average of the average value of the N total movement commands in a loop processing cycle is used. 4. A differential value of the average value in the position / velocity loop processing advanced by a set number of cycles from the position / velocity loop period is obtained, and the differential value is obtained by multiplying the differential value by a velocity feedforward coefficient. 4. The feedforward control method for a servomotor according to claim 1, wherein the speed feedforward amount is added to a torque command value obtained by the speed loop processing to obtain a torque command to the servomotor.