JP4496545B2 - Digital servo control device and control method thereof - Google Patents

Description

  The present invention relates to a digital servo control device that uses torque feedforward and has improved positioning performance during digital control, and a control method therefor.

In the conventional servo control method, the position command is differentiated into a speed feedforward signal, differentiated again to create a torque feedforward signal, and used as it is. (For example, refer to Patent Document 1).
FIG. 5 is a block diagram of a conventional servo control device. In FIG. 5, KP of the transfer function 30 is a position gain in the position loop, K1 of the transfer function 32 is an integral gain in the speed loop, k2 of the transfer function 34 is a proportional gain in the speed loop, 36 is a current loop circuit, and 38 is a servo. The electric part of the motor, R is the resistance of the winding, L is the inductance of the winding, 40 is the mechanical part of the servo motor, Kt is the torque constant, Jm is the inertia, 42 is the rotation speed of the servo motor, and the position Is a transfer function for calculating. The transfer function 44 is a position feedforward term, and α1 is a position feedforward coefficient. 46 is a speed feedforward term, and α2 is a speed feedforward coefficient. The speed feedforward coefficient α2 is normally a value close to the value of Jm / Kt (Jm: inertia, Kt: torque constant). Note that the value of the feedforward coefficient α1 of the position feedforward term 44 is experimentally determined according to the characteristics of the motor and the like (ideally “1” is good).

In the conventional control, the position command a is differentiated, and the position feedforward coefficient α1 is multiplied by the differential value to obtain a position feedforward control amount. Thus, the normal position loop control, that is, the current position P of the motor from the position command a. To obtain a position deviation ε, and multiply this by a position loop gain KP to obtain a normal speed command. Then, a position feedforward control amount is added to this normal speed command to obtain a position command Vc for which position feedforward control has been performed.
On the other hand, the position feedforward control amount is differentiated, and the speed feedforward coefficient α2 is multiplied to obtain the speed feedforward control amount, and the speed loop control (IP control), that is, the servo motor actuality is determined from the speed command Vc. By subtracting the value obtained by subtracting the proportional gain k2 from the actual speed V of the servo motor from the value obtained by subtracting the speed V to obtain the speed deviation, integrating the speed deviation and multiplying by the integral gain k1, The speed feedforward control amount is added to obtain a current command Ic.

  If the value of the position command a changes, the position deviation ε also increases and the speed command output in the normal position loop processing also changes greatly, but there is a delay in the position loop processing. However, the position feedforward control also increases the position feedforward control amount according to the amount of change in the position command a and adds it to the speed command, resulting in a feedforward controlled speed command. Compensated. The same applies to the speed loop, and the current command by the normal speed loop process also changes according to the change of the speed command, but there is a delay because of the integral term. However, in this case as well, since the speed feedforward control amount is added by the speed feedforward control and becomes a current command, the delay of the speed loop is also compensated, and the response of the servo system is improved as a whole. As a result, the follow-up performance of the servo motor with respect to the position command a is improved, and the undulation of the position deviation is reduced.

When the conventional control is realized by digital control, the following method has been adopted.
Suppose that the cycle of position and velocity loop processing is TP, and the position command in each position and velocity loop processing is a (n) (n = 1, 2, 3..., A (n) = 0 when n ≦ 0). The differential value b (n) of the position command a (n) is actually calculated as a difference by the following equation (1).

  b (n) = {a (n) -a (n-1)} / Tp (1)

  The position feedforward signal is obtained by multiplying the value of b (n) by the position feedforward coefficient α1 to obtain a feedforward control amount c (n).

  c (n) = α1 · b (n) (2)

  The speed feedforward signal Dn is obtained by subtracting the feedforward control amount c (n-1) at the position of the previous period from the position feedforward control amount c (n) as shown in the equation (3). Multiplied by (α2 / TP).

  d (n) = α2 · {(c (n) −c (n−1)} / Tp (3)

In this way, in the conventional feedforward control device, the position command is simply differentiated (the difference between the current value and the previous value divided by the sampling period) and multiplied by α1 to obtain the velocity feedforward signal and differentiated again. A procedure has been taken in which a product of α2 multiplied by α2 is used as a current (or torque) feedforward signal.
Japanese Patent No. 2762364 (page 4-7, FIG. 1, FIG. 8)

In the conventional servo control method, the position command is simply differentiated (the difference between the current value and the previous value divided by the sampling period) multiplied by α1 to be the speed feedforward signal, and the derivative once again multiplied by α2. Therefore, the current (or torque) feed-forward signal is used, and ideally, the speed feed-forward signal is 0 for the next one sampling after the speed command is completed. Although the forward signal needs to be 0, the torque feed forward signal is output by one sampling as shown by the hatched portion in FIG. 3B due to the 0th-order hold, resulting in overcompensation. Thus, there is a problem in that overshoot and an increase in position deviation are caused.
In addition, since there is actually one sampling delay in the control calculation, a torque feed forward signal may be output for two samplings as shown by the hatched portion in FIG. In addition to the delay, an additional 0.5 sampling time delay occurs due to the difference approximation, so that the torque feedforward signal is output by an extra 2.5 sampling amount as shown by the hatched portion in FIG. As a result, there is a problem that over-compensation is caused, resulting in an increase in overshoot and position deviation.

  The present invention has been made in view of such problems, prevents overcompensation of the torque feedforward signal, has a very small deviation at the end of acceleration / deceleration and at the time of positioning, and is easy to calculate. It is an object of the present invention to provide a digital servo control device and a control method thereof that can cope with the above.

In order to solve the above problem, the present invention is as follows.
According to a first aspect of the present invention, there is provided a control method for a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward generation unit, and a torque feedforward generation unit. Using the incremental value dva and the maximum velocity vmax of the speed command at the first sampling, the sampling kaend at which the acceleration ends is obtained by calculation, and the detected value that the controlled object is operated by the torque command output from the torque command delivery is obtained. When the delay until acquisition is s sampling, processing is performed by a procedure of forcibly setting the torque feedforward signal during acceleration after (kaend-s) sampling to 0.

  According to a second aspect of the present invention, there is provided a control method for a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward generation unit, and a torque feedforward generation unit. Using the decrement value dvd of the speed command at the time of the first sampling and the speed vd at the start of deceleration, the sampling kend at which the deceleration ends is obtained by calculation, and the control target is operated by the torque command output from the torque command delivery When the delay until the detection value is obtained is s sampling, the processing is performed by the procedure of forcibly setting the torque feedforward signal at the time of deceleration after (kdend-s) sampling to 0. is there.

  According to a third aspect of the present invention, there is provided a control method for a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feed forward generation unit, and a torque feed forward generation unit. Using the incremental value dva and the maximum velocity vmax of the speed command at the first sampling, the sampling kaend at which the acceleration ends is obtained by calculation, and the detected value that the controlled object is operated by the torque command output from the torque command delivery is obtained. When the delay until acquisition is s sampling, the torque feedforward signal during acceleration after (kaend-s) sampling is forcibly set to 0, and the decrement value dvd of the speed command at the first sampling of deceleration start and Using the velocity vd at the start of deceleration, the sampling kend at which deceleration ends is calculated, When s sampling is used as the delay until the detection value obtained when the control target is operated by the torque command output from the torque command payout, the torque feed forward signal at the time of deceleration after (kend-s) sampling is forcibly set to 0. It is characterized by processing according to the procedure.

  According to a fourth aspect of the present invention, in the control method of the digital servo control device according to the first or third aspect, the following calculation formula is used in the calculation for obtaining the sampling kaend at which the acceleration ends. Is.

kaend = k0 + vmax / dva-1
k0: Acceleration start command sampling

According to a fifth aspect of the present invention, in the control method of the digital servo control device according to the second or third aspect, the following calculation formula is used in the calculation for obtaining the sampling kend at which the deceleration ends. Is.
kdend = n0 + vd / dvd−1
n0: Sampling command start sampling

  According to a sixth aspect of the present invention, in the control method of the digital servo control device according to any one of the first to third aspects, the value of the maximum speed vmax is acquired from a host device at the start of command payout. Is.

  According to a seventh aspect of the present invention, in the control method of the digital servo control device according to any one of the first or third aspects, a detection value obtained by operating the control target by the torque command output from the torque command payout is acquired. When the delay s is a fractional number, when sampling with the decimal part of kaend-s rounded down, the torque feedforward value signal at acceleration is set to an appropriate integer α, and when sampling with the decimal part rounded up, acceleration The torque feed forward signal is set to 0, and the process is performed.

  According to an eighth aspect of the present invention, in the control method of the digital servo control device according to the seventh aspect, the appropriate integer α is set to a value obtained by rounding down the decimal point of the reciprocal of the value after the decimal point of kaend-s. It is a feature.

  According to a ninth aspect of the present invention, in the control method of the digital servo control device according to any one of the second or third aspects, a detection value obtained by operating a controlled object by a torque command output from the torque command payout is acquired. When the delay s is a fractional number, when sampling is performed with the decimal point of kend-s rounded down, the torque feedforward value signal during deceleration is set to an appropriate integer α, and when sampling is rounded up after decimal point, during deceleration The torque feed forward signal is set to 0, and the process is performed.

  According to a tenth aspect of the present invention, in the control method of the digital servo control device according to the ninth aspect, the appropriate integer α is set to a value obtained by rounding down the decimal part of the reciprocal of the kend-s. It is characterized by this.

  According to the eleventh aspect of the present invention, there is provided a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feed forward generation unit, and a torque feed forward generation unit. The control target is operated by the torque command output from the acceleration command issuing unit, and the acceleration command predicting unit that calculates the sampling kaend at which the acceleration is completed by using the speed command increment dva and the maximum velocity vmax at the time of sampling And (kaend-s) a torque feedforward signal correction unit that forcibly sets the torque feedforward signal during acceleration after sampling to s sampling when the delay until the detected value is acquired is s sampling. It is a feature.

  According to a twelfth aspect of the present invention, there is provided a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feed forward generation unit, and a torque feed forward generation unit. Controlled by the torque command output from the deceleration command prediction unit for calculating the sampling kend at which deceleration ends by using the decrement value dvd of the speed command at the time of sampling and the speed vd at the start of deceleration A torque feedforward signal correction unit that forcibly sets a torque feedforward signal during deceleration after sampling (kend-s) to s sampling when the delay until the detection value of the target is acquired is s sampling; It is characterized by having.

  According to a thirteenth aspect of the present invention, in the digital servo control device including the position control hand portion, the speed control portion, the current control portion, the speed feedforward creation portion, and the torque feedforward creation portion, the first command payout start An acceleration end prediction unit that calculates a sampling kaend at which acceleration ends by using an increment value dva and a maximum speed vmax at the time of sampling, and a speed command decrement value dvd at the first sampling of deceleration start And a deceleration end prediction unit that calculates a sampling kend at which deceleration ends by using the velocity vd at the time of deceleration start, and from the torque command delivery until a detection value obtained by operating the control target by the output torque command is obtained. When s sampling is used as the delay, the torque fee during acceleration after (kaend-s) sampling A torque feedforward signal correction unit for forcibly setting the forward signal to 0 and forcibly setting the torque feedforward signal during deceleration after (kdend-s) sampling to 0 It is.

  According to the fourteenth aspect of the present invention, in the digital servo control device including the speed control unit, the current control unit, and the torque feedforward creation unit, the speed command increment dva and maximum Deceleration is performed using the acceleration end prediction unit for calculating the sampling kaend at which acceleration ends by using the speed vmax, the speed command decrement value dvd at the first sampling of deceleration start, and the speed vd at the start of deceleration. When the s-sampling is the delay from the end of the deceleration command to obtain the detected value that the controlled object is operated by the output torque command from the torque command payout, s) The torque feedforward signal during acceleration after sampling is forcibly set to 0, and (kd end-s) including a torque feedforward signal correction unit that forcibly sets the torque feedforward signal during deceleration after sampling to zero.

  According to a fifteenth aspect of the present invention, there is provided a control method for a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward generation unit, and a torque feedforward generation unit. When the sampling kend at the time and the sampling kend at the end of deceleration are received from the host device, and the delay until the detection value obtained when the control target is operated by the torque command output from the torque command payout is s sampling, (kaend -S) The torque feed forward signal during acceleration after sampling is forcibly set to 0, and (kdend-s) The torque feed forward signal during deceleration after sampling is forcibly set to 0. It is characterized by this.

  According to a sixteenth aspect of the present invention, in the control method of the digital servo control device according to the fifteenth aspect, instead of directly receiving the sampling kaend at the end of the acceleration and the sampling kend at the end of the deceleration, the time katime at which the acceleration ends and The value of the time kdtime at which deceleration ends is received from the host device, and these values are divided by the sampling period Ts to calculate the sampling kaend at the end of acceleration and the sampling kend at the end of deceleration.

According to the first and eleventh aspects of the present invention, the acceleration end point can be accurately determined, and an excessive torque feedforward signal is not output at that time. Therefore, overcompensation can be avoided, and the deviation at the end of acceleration is reduced. Can be small.
Further, according to the invention described in claim 2 and claim 12, the end point of deceleration can be accurately known, and an excessive torque feedforward signal is not output at that time, so that overcompensation can be avoided, and at the end of deceleration. Deviation can be reduced.
In addition, according to the invention described in claim 3 and claim 13, the acceleration end point and the deceleration end point are accurately known, and an excessive torque feedforward signal is not output at that time, so that overcompensation can be avoided. Deviations at both the end of acceleration and the end of deceleration can be reduced.
According to the fourth aspect of the present invention, at the time of a command for constant acceleration such as a trapezoidal command, the acceleration end point can be predicted by simple calculation, and the deviation at the end of acceleration can be reduced.
According to the fifth aspect of the present invention, when a command for constant deceleration such as a trapezoid command is given, the end point of deceleration can be predicted by simple calculation, and the deviation at the end of deceleration can be reduced.
According to the invention described in claim 6, since vmax is acquired from the host device, the present invention can be applied to any command even if the command maximum speed vmax is not known in advance. .
According to the seventh aspect of the present invention, even when the delay s until the detection value obtained by operating the control target by the torque command output from the torque command payout is a small number, an excessive torque feed forward at that time Since no signal is output, overcompensation can be avoided and deviation at the end of acceleration can be reduced.
According to the ninth aspect of the present invention, even when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is a small number, an excessive torque feed forward at that time Since no signal is output, overcompensation can be avoided and deviation at the end of deceleration can be reduced.
According to the invention described in claims 8 and 10, when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is a decimal, the value of the integer α is set. Since the reciprocal of the value after the decimal point of kend-s is set, α can be easily set, and even when s is a decimal, the deviation between the end of acceleration and the end of deceleration can be easily reduced.
According to the invention described in claim 14, even when the digital control device performs speed control, the deviation between the end of acceleration and the end of deceleration can be reduced.
Further, according to the inventions of claims 15 and 16, since information corresponding to the sampling kend at the end of acceleration and the sampling kend at the end of deceleration can be obtained directly, the difference between the end of acceleration and the end of deceleration can be obtained with a little calculation. Can be reduced.

  Hereinafter, specific examples of the method of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a digital servo control apparatus that implements the method of the present invention. In the figure, reference numeral 1 denotes a control object, and 2 denotes a detector that detects the position and speed of the control object. Reference numeral 100 denotes a digital servo control device of the present invention. The position command ref and the position detection value x detected by the detector 2 attached to the controlled object are input, the control calculation is digitally calculated, and the current I is output to the motor.
Reference numeral 3 represents a position control unit, 4 represents a speed control unit, and 5 represents a current control unit. Each control unit performs proportional or proportional-integral control. Reference numeral 6 denotes a speed feedforward creating unit that differentiates the position command and creates a speed feedforward signal vff. Reference numeral 7 denotes a torque feedforward creation unit, which further differentiates the speed feedforward signal vff and multiplies the inertia to be controlled to create a torque feedforward signal tff.
Reference numeral 8 denotes an acceleration end prediction unit, which receives vmax and a command from the host device, calculates how many times the acceleration ends when counting from the sampling at the start of command delivery, and outputs a sampling kaend at the end of acceleration. . If the command is already known, the value of vmax may be stored in advance in the memory instead of being obtained from the host device.
Reference numeral 9 denotes a deceleration end prediction unit, which calculates from the speed vd at the start of deceleration and the command, how many samplings from the deceleration start sampling to calculate when deceleration will end, and outputs the sampling kend at the end of deceleration. .
10 represents a torque feedforward signal correction unit. When s sampling is used as the delay from the torque command delivery until the detection value obtained when the control target is operated by the output torque command is obtained, s, kaend, kend, and tff are set as follows. The torque feed forward signal tff is corrected so that the torque feed forward signal is not overcompensated. Here, s may be stored in the memory in advance, or may be adjusted as a parameter. Reference numerals 11 to 13 denote arithmetic units.

Next, a digital calculation method for each unit will be described.
Hereinafter, the sampling time is defined as Ts, the current sampling is defined as kth, the uth previous sampling is defined as ku, and the value at the time of the kuth sampling of the variable q is expressed as q (ku). To.

  First, the position control unit 3 performs a calculation by multiplying the difference between the position command ref (k) and the position detection value x (k) by the position loop gain Kp and outputs the speed command vref (k) as shown in Expression (4). To do.

    vref (k) = Kp · {ref (k) −x (k)} (4)

  Next, in the speed control unit 4, a speed feedforward signal vff (k), which is an output of the speed FF creation unit, is added to the difference between the speed command vref (k) and the speed detection value v (k), as shown in Expression (5). ) Is multiplied by the velocity loop proportional gain Kv to calculate sref (k).

    sref (k) = Kv · {vref (k) −v (k) + vff (k)} (5)

Here, v (k) may be obtained by differentiation by differential approximation of the position detection value x (k).
Next, as shown in equation (6), sref (k) is multiplied by the integral control gain Ki and sampling time Ts, and added to the previous value to perform an integral operation to calculate si (k).

    si (k) = si (k−1) + Ts · Ki · sref (k) (6)

  Next, as shown in Expression (7), the torque command tref (k) is calculated by adding sref (k) and si (k).

    tref (k) = sref (k) + si (k) (7)

  Next, the current control unit 5 inputs the sum of the torque command tr (k) and the torque feedforward signal tff (k), performs unit conversion and control calculation, and calculates the current value I (k). Here, the processing inside the current control unit is actually complicated processing such as conversion of alternating current to direct current, but the processing method inside the current control means has nothing to do with the present invention, and what kind of processing is performed? The description is omitted here.

  The speed FF creating unit 6 differentiates the position command ref (k) by difference approximation as shown in the equation (8), and calculates a speed feedforward signal vff (k).

    vff (k) = {ref (k) −ref (k−1)} / Ts (8)

  In the torque FF creation unit 7, as shown in equation (9). The torque feedforward signal tff (k) is calculated by differentiating the speed feedforward signal vff (k) by differential approximation and multiplying by the inertia Jn.

    tff (k) = Jn · {vff (k) −vff (k−1)} / Ts (9)

  As shown in equation (10), the acceleration end time prediction unit 8 obtains an incremental value dva of the sampling speed at the k0th sampling at the start of command payout, and receives vmax received from the higher order commander as shown in equation (11). And the sampling kaend at the end of acceleration are output from dva.

dva = {ref (k0) -ref (k0-1)} / Ts
-{Ref (k0-1) -ref (k0-2)} / Ts (10)

  kaend = k0 + vmax / dva-1 (11)

Furthermore, it demonstrates in detail using FIG.
FIG. 3A shows the command speed obtained by differentiating the position command input from the upper level by difference approximation. k0 represents sampling at the start of the command. In this embodiment, the maximum speed vmax is four times as large as dva. Therefore, from equation (11):

kaend = k0 + 4−1 = k0 + 3 (12)
Thus, the sampling at the time when the acceleration ends is k0 + 3.

The torque FF signal correction unit 10 performs a process for forcibly setting the torque feed forward after sampling (kaend-s) to zero.

  In FIG. 3 (b), the hatched portion is an extra torque feed forward signal when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is set to 0 sampling. is there. in this case,

kaend-s = k0 + 3-0 = k0 + 3
Subsequent torque feed forward signals (torque areas) are forced to zero.

  In FIG. 3 (c), the hatched portion is an extra torque feedforward signal when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is 1 sampling. is there. in this case,

kaend-s = k0 + 3-1 = k0 + 2
Subsequent torque feed forward signals (torque areas) are forced to zero.

In FIG. 3 (d), the shaded portion indicates an extra torque feed forward when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is 1.5 samplings. Signal. in this case,

kaend-s = k0 + 3-1.5 = k0 + 1.5
Subsequent torque feed forward signals (torque areas) are forced to zero.

  Thus, when the calculation result is not an integer, the torque feedforward signal is set to 1 / α at the time of sampling with the decimal part of kaend-s rounded down, and the torque feedforward signal is set to 0 at the time of sampling with the decimal part rounded up. This can be done.

  That is, in the case as shown in FIG. 3D, the torque feed forward signal is set to 1 / α for the k0 + 1 sampling with the decimal part rounded down to 1.5, and the torque feed for the sampling of k0 + 2 with the decimal part rounded up. The forward signal may be set to 0.

  Α may be obtained by adjustment, but may be an inverse number of a value after the decimal point. That is, in the case as shown in FIG. 3D, since the value after the decimal point is 0.5, α = 1 / 0.5 = 2.

  Next, processing during deceleration will be described. The process at the time of deceleration is basically the same method as the process at the time of acceleration described above.

  The deceleration end time prediction unit 9 obtains a decrement value dvd of the sampling speed at the start of deceleration n0 as shown in Expression (13), and calculates the speed vd at the start of deceleration as shown in Expression (14). The sampling kend at the end of deceleration is output from dvd.

dvd = {ref (n0) -ref (n0-1)} / Ts
-{Ref (n0-1) -ref (n0-2)} / Ts (13)

  kdend = n0 + vd / dvd−1 (14)

  Furthermore, it demonstrates in detail using FIG. FIG. 3A shows the command speed obtained by differentiating the position command input from the upper level by difference approximation. n0 represents sampling at the start of deceleration. The speed vd (= vmax) at the start of deceleration is four times as large as dvd. Therefore, from equation (14),

kdend = n0 + 4−1 = n0 + 3 (15)
Thus, sampling at the time when deceleration ends is n0 + 3.
The torque FF signal correction unit 10 performs a process for forcibly setting the torque feed forward after (kend−s) to zero.

  In FIG. 3 (b), the hatched portion is an extra torque feed forward signal when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is set to 0 sampling. is there. in this case,

kdend-s = n0 + 3-0 = n0 + 3
Subsequent torque feed forward signals (torque areas) are forced to zero.

  In FIG. 3 (c), the hatched portion is an extra torque feedforward signal when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is 1 sampling. is there. in this case,

kdend-s = n0 + 3-1 = n0 + 2
Subsequent torque feed forward signals (torque areas) are forced to zero.

  In FIG. 3 (d), the shaded portion indicates an extra torque feed forward when the delay s until the detection value obtained by operating the controlled object by the torque command output from the torque command payout is 1.5 samplings. Signal. in this case,

kdend-s = n0 + 3-1.5 = n0 + 1.5
Subsequent torque feed forward signals (torque areas) are forced to zero.

  As described above, when the calculation result is not an integer, the torque feed forward value signal is set to 1 / α at the time of sampling with the fractional part of the knd-s rounded down, and the torque feed forward signal is set to 0 at the time of sampling with the decimal part rounded up. It can respond by doing.

  That is, in the case of FIG. 3D, the torque feed forward signal is set to 1 / α for the (n0 + 1) th sampling with the fractional part of 1.5 rounded down, and the torque feed is used for the sampling of n0 + 2 with the fractional part rounded up. The forward signal may be set to 0.

Α may be obtained by adjustment, but may be an inverse number of a value after the decimal point. That is, in the case as shown in FIG. 3D, since the value after the decimal point is 0.5, α = 1 / 0.5 = 2.
The above is the description of the first embodiment of the present invention.

FIG. 3 is a simulation result showing the effect of the present invention. FIG. 3A shows waveforms of speed (upper stage) and deviation (lower stage) during conventional operation without using the present invention. As can be seen from the figure, a large deviation occurs due to overcompensation by the torque feedforward signal at the end of acceleration and at the end of deceleration.
FIG. 3B shows waveforms of speed (upper stage) and deviation (lower stage) when the present invention is used. As can be seen from the figure, since the over-compensation by the torque feedforward signal does not occur at the end of acceleration and at the end of deceleration, the deviation is almost zero.

  In this way, by using the speed command increment dva and the maximum speed vmax at the time of the first sampling of command payout, the sampling kaend at which the acceleration ends is calculated, and control is performed from the torque command payout by the output torque command. When s sampling is used as the delay until the detection value obtained by the operation of the target is s sampling, the torque feed forward signal after the (kaend-s) sampling is forcibly set to 0, resulting in torque feed forward. The signal is not overcompensated, and the deviation at the end of acceleration and positioning can be made very small.

  FIG. 2 is a block diagram showing the configuration of a digital servo control apparatus that implements the second method of the present invention. The difference between the present embodiment and the first embodiment is that in the first embodiment, the position control is performed with respect to the position command by 100 digital servo control devices, whereas in the present embodiment, the speed control is performed with respect to the speed command. Is about to do. Since the basic method is the same as that of the first embodiment, only the parts different from the first embodiment will be described below.

  As shown in equation (16), the acceleration end time prediction unit 8 obtains an incremental value dva of the sampling speed at the k0th sampling at the start of command payout, and receives vmax received from the higher order command device as shown in equation (17). And the sampling kaend at the end of acceleration are output from dva. Since the command input in advance is the speed command vref (k), the calculation of dva is different from that in the first embodiment.

  dva = {vref (k0) -vref (k0-1)} (16)

  kaend = k0 + vmax / dva-1 (17)

  The deceleration end prediction means 9 obtains the decrement value dvd of the speed of the n0th sampling portion at the start of deceleration as shown in the equation (18), and the velocity vd at the start of deceleration as shown in the equation (19). Sampling kend at the end of deceleration is output from (= vmax) and dvd.

  dvd = {vref (n0) -vref (n0-1)} (18)

kdend = n0 + vd / dvd−1 (19)
The torque FF signal correction unit 10 performs processing for forcibly setting the torque feedforward signal after sampling (kaend-s) and (kdnd-s) to zero.
The above is the description of the second embodiment of the present invention. In the case of this configuration, the same effect as in the first embodiment is obtained.

  Using the speed command increment dva and the maximum speed vmax at the time of the first sampling of command payout, the sampling kaend at which the acceleration ends is calculated, and the control target is operated by the torque command output from the torque command payout When the delay until the detection value is acquired is s sampling, the process is performed in such a manner that the torque feed forward signal during acceleration after (kaend-s) sampling is forcibly set to 0. Since compensation can be prevented and the deviation at the time of completion of acceleration and positioning can be made extremely small, it can be applied not only to single-axis mechanisms but also to multi-axis mechanism positioning such as robots and machine tools.

The block diagram which shows the structure of the digital servo control apparatus which applies the method of this invention Block diagram showing the configuration 2 of the digital servo control apparatus to which the method of the present invention is applied. The figure explaining the effect | action of this invention Figure of simulation results of the present invention Block diagram showing the configuration of a conventional digital servo controller

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Control object 2 Detector 3 Position control part 4 Speed control part 5 Current control part 6 Speed FF creation part 7 Torque FF creation part 8 Acceleration end prediction part 9 Deceleration end prediction part 10 Torque FF signal correction part 30 Position loop 32 Speed loop integration element 34 Speed feedback gain 36 Current loop 38 Servo motor electrical section 40 Servo motor machine section 42 Integration 44 Position feed forward term 46 Speed feed forward term 100 Digital servo controller

Claims (16)

In a control method of a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward creation unit, and a torque feedforward creation unit,
Using a speed command increment dva and a maximum speed vmax at the time of sampling for the first time of command payout start, a sampling kaend at which acceleration ends is obtained by calculation,
When s sampling is used as the delay until the detection value obtained when the controlled object is operated by the torque command output from the torque command is issued, the torque feed forward signal during acceleration after (kaend-s) sampling is forcibly set to 0. And
A control method for a digital servo control device, characterized in that processing is performed according to the following procedure. In a control method of a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward creation unit, and a torque feedforward creation unit,
Using the decrement value dvd of the speed command at the first sampling of deceleration start and the speed vd at the start of deceleration, a sampling kend at which deceleration ends is calculated,
When s sampling is used as the delay until the detection value obtained when the control target is operated by the torque command output from the torque command payout, the torque feed forward signal at the time of deceleration after (kend-s) sampling is forcibly set to 0. And
A control method for a digital servo control device, characterized in that processing is performed according to the following procedure. In a control method of a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward creation unit, and a torque feedforward creation unit,
Using a speed command increment dva and a maximum speed vmax at the time of sampling for the first time of command payout start, a sampling kaend at which acceleration ends is obtained by calculation,
When s sampling is used as the delay until the detection value obtained when the controlled object is operated by the torque command output from the torque command is issued, the torque feed forward signal during acceleration after (kaend-s) sampling is forcibly set to 0. age,
Using the decrement value dvd of the speed command at the first sampling of deceleration start and the speed vd at the start of deceleration, a sampling kend at which deceleration ends is calculated,
When s sampling is used as the delay until the detection value obtained when the control target is operated by the torque command output from the torque command payout, the torque feed forward signal at the time of deceleration after (kend-s) sampling is forcibly set to 0. And
A control method for a digital servo control device, characterized in that processing is performed according to the following procedure. 4. The method of controlling a digital servo control device according to claim 1, wherein the following calculation formula is used in the calculation for obtaining the sampling kaend at which the acceleration ends.
kaend = k0 + vmax / dva-1
k0: Acceleration start command sampling 4. The control method for a digital servo control device according to claim 2, wherein the calculation for obtaining the sampling kend at which the deceleration ends is calculated using the following formula.
kdend = n0 + vd / dvd−1
n0: Sampling command start sampling   4. The control method for a digital servo control device according to claim 1, wherein the value of the maximum speed vmax is acquired from a host device at the start of command payout.   When the delay s until the detection value obtained by operating the control object according to the torque command output from the torque command payout is a decimal number, the value of torque feed forward at the time of acceleration when sampling is performed with the decimal part of kaend-s rounded down. 4. The signal is processed by the procedure of setting the signal to an appropriate integer [alpha], and setting the torque feedforward signal during acceleration to 0 at the time of sampling rounded up after the decimal point. Method of digital servo control device.   8. The control method for a digital servo control device according to claim 7, wherein the appropriate integer [alpha] is a value obtained by rounding down the decimal part of the reciprocal of kaend-s.   When the delay s until the detection value obtained when the controlled object is operated by the torque command output from the torque command payout is a fractional value, the value of torque feed forward at the time of deceleration at the time of sampling with the decimal point of kend-s rounded down 4. The process according to claim 2, wherein the signal is set to an appropriate integer [alpha] and the torque feed forward signal at the time of deceleration is set to 0 at the time of sampling rounded up after the decimal point. Method of digital servo control device.   10. The control method for a digital servo control device according to claim 9, wherein the appropriate integer [alpha] is a value obtained by rounding down the decimal point of the reciprocal of the kend-s. In a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward creation unit, and a torque feedforward creation unit,
An acceleration end prediction unit for calculating a sampling kaend at which acceleration ends by using an increment value dva and a maximum speed vmax of a speed command at the time of first sampling of command payout;
When s sampling is used as the delay from the torque command delivery until the detection value obtained when the controlled object is operated by the output torque command, the torque feed forward signal during acceleration after sampling is forced A torque feedforward signal correction unit for setting zero,
A digital servo control device comprising: In a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward creation unit, and a torque feedforward creation unit,
A deceleration end prediction unit for calculating a sampling kend at which deceleration ends by using a decrement value dvd of the speed command at the first sampling of deceleration start and a speed vd at the start of deceleration;
When s sampling is used as the delay from the torque command delivery until the detection value obtained when the controlled object is operated by the output torque command is taken, the torque feed forward signal at the time of deceleration after (kend-s) sampling is forced A torque feedforward signal correction unit for setting zero,
A digital servo control device comprising: In the digital servo control device including the position control hand part, the speed control part, the current control part, the speed feed forward creation part, and the torque feed forward creation part,
An acceleration end prediction unit for calculating a sampling kaend at which acceleration ends by using an increment value dva and a maximum speed vmax of a speed command at the time of first sampling of command payout;
A deceleration end prediction unit for calculating a sampling kend at which deceleration ends by using a decrement value dvd of the speed command at the first sampling of deceleration start and a speed vd at the start of deceleration;
When s sampling is used as the delay from the torque command delivery until the detection value obtained when the controlled object is operated by the output torque command, the torque feed forward signal during acceleration after sampling is forced A torque feedforward signal correction unit that forcibly sets the torque feedforward signal during deceleration after the sampling to zero and (kdend-s) sampling to 0;
A digital servo control device comprising: In the digital servo control device provided with the speed control unit, the current control unit, and the torque feed forward creation unit,
An acceleration end prediction unit for calculating a sampling kaend at which acceleration ends by using an increment value dva and a maximum speed vmax of a speed command at the time of first sampling of command payout;
A deceleration end prediction unit for calculating a sampling kend at which deceleration ends by using a decrement value dvd of the speed command at the first sampling of deceleration start and a speed vd at the start of deceleration;
When s sampling is used as the delay from the torque command delivery until the detection value obtained when the controlled object is operated by the output torque command, the torque feed forward signal during acceleration after sampling is forced A torque feedforward signal correction unit that forcibly sets the torque feedforward signal during deceleration after the sampling to zero and (kdend-s) sampling to 0;
A digital servo control device comprising: In a control method of a digital servo control device including a position control unit, a speed control unit, a current control unit, a speed feedforward creation unit, and a torque feedforward creation unit,
When the sampling kend at the end of the acceleration and the sampling kend at the end of the deceleration are received from the host device, and the delay until the detection value that the control target is operated by the torque command output from the torque command payout is s sampling, (Kaend-s) The torque feedforward signal during acceleration after sampling is forcibly set to 0, and the torque feedforward signal during deceleration after sampling (kend-s) is forcibly set to 0.
A control method for a digital servo control device, characterized in that processing is performed according to the following procedure.   Instead of directly receiving the sampling kend at the end of the acceleration and the sampling kend at the end of the deceleration, the value of the time katime at which the acceleration ends and the value of the time kdtime at which the deceleration ends are received from the host device and divided by the sampling period Ts. 16. The control method of a digital servo control device according to claim 15, wherein a sampling kend at the end of acceleration and a sampling kend at the end of deceleration are calculated.

Images