JPH10225157A - Numeric controller with deviation counter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance stability during a stop by interposing a function calculator between a deviation counter and a DC converter and providing a servo motor with a speed command determined according to a specified calculation formula thereby shortening the settling time. SOLUTION: A function calculator 12, comprising a microcomputer or a signal processor, is interposed between a deviation counter 3 and a DC converter 5. The function calculator 12 receives the count 4 of the deviation counter 3 and delivers a speed command value 13, calculated according to an expression V=K.Z<n> (n=0.7-0.9) to the DC converter 5. In the expression, V: speed command value, K: proportional constant (the speed command value is proportional to the rotational speed of a servo motor when time lag is neglected), Z: count of the deviation counter). Since the settling time can be shortened, a high speed machine tool or industrial robot employing a numeric controller with a deviation counter can be realized.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

2. Description of the Related Art Numerical controllers using a deviation counter are generally used for controlling a stop position of a moving distance or a rotation angle of a machine tool, an industrial robot or the like.

[0002]

2. Description of the Related Art FIG. 1 is a block diagram of a known numerical controller using a conventional deviation counter. Referring to FIG. 1, 1 is an oscillation circuit, 2 is an oscillation pulse output from the oscillation circuit, and 3 is a deviation counter. The deviation counter 3 adds and counts the oscillation pulse 2 output from the oscillation circuit 1. 4 is the count value of the deviation counter. 5 is a D / A converter, and 6 is a speed command. D
The / A converter 5 receives the count value 4 of the deviation counter 3, converts it into a voltage proportional to the count value 4, and converts it into a speed command 6. 7 is a servo driver, 8 is a motor, and 9 is the power of the motor. Reference numeral 10 denotes a rotary encoder, which is connected to the rotation shaft of the motor 8 and generates, for example, 1000 pulses per rotation in accordance with the rotation angle of the rotation shaft of the motor 8. Reference numeral 11 denotes a pulse generated by the rotary encoder 10. The servo driver 7 inputs the generated pulse 11 as feedback of the rotation speed of the motor 8 and controls the power 9 so that the rotation speed of the motor 8 becomes a rotation speed corresponding to the speed command value 6. A set of devices combining the servo driver 7, the motor 8, the power 9, the rotary encoder 10, and the like is called a servo motor. The deviation counter 3 counts down the number of generated pulses 11. With this structure, the deviation counter 3 outputs a count value 4 of a difference between the number of oscillation pulses of the oscillation circuit 1 and the number of pulses generated by the rotary encoder 10. The motor rotates at a rotation speed corresponding to the count value 4 of the deviation counter. The solid line in FIG. 2 indicates the pulse of the oscillation pulse 2 per second. The dotted line in FIG. 2 indicates the number of generated pulses 11 per second. FIG. 2 shows time on the horizontal axis and pulse per second on the vertical axis. In FIG. 2, a point p is a point where the oscillation of the oscillation circuit 1 stops, and a point q is a point where the generated pulse 11 of the rotary encoder 10 stops, that is, a stop point of the servomotor. In a numerical control device using a deviation counter, the time from the point p at which the oscillation of the oscillation circuit 1 stops to the point q at which the rotation of the servomotor stops is referred to as a stop stabilization time. The stop stabilization time usually requires 0.1 seconds, and at least 0.05 seconds or more is required. Therefore, it is a serious drawback as a numerical control device for a mechanical device that operates at a high frequency. As a conventional technique, various cam curves have been proposed as speed curves for moving an object at high speed. Since the rotation speed control of the servo motor is a feedback control, there is a time delay from when a speed command is given to the servo motor to when the servo motor reaches a rotation speed corresponding to the speed command. In order to stably stop the servomotor with a delay time in a short time, a speed curve with a small change in acceleration, which sufficiently reduces the rate of change in acceleration at the stop point, is appropriate.

[0003]

SUMMARY OF THE INVENTION The present invention provides a numerical control device for a machine tool or an industrial robot which operates frequently by shortening the stop stabilization time.
An effective numerical control device is provided. The present invention
By using an effective speed curve, it is possible to shorten the stop stabilization time and provide a numerical control device with good stability at the stop.

[0004]

FIG. 3 shows various curves of the rate of change of acceleration, acceleration, and speed from point p to point q in FIG. The curve A shows the theoretical value of a conventional curve using a deviation counter. Although the stability at the time of stopping is good, there is a disadvantage that the stopping stabilization time becomes redundant. B
The curve is a curve to be implemented by the present invention, and is a speed curve in which the rate of change in acceleration decreases toward the stop point and the rate of change in acceleration at the stop point becomes zero. The stability at the time of stop is good, and the stop stabilization time can be shortened. C
The curve is a curve in which the rate of change in acceleration is constant. Since the rate of change in acceleration at the stop point does not become 0, the stability at the time of stopping is not good. The curve D in FIG. 3 shows a curve at the time of deceleration of a deformed trapezoidal curve or a deformed sine curve generally adopted as a cam curve. Since the rate of change in acceleration at the stop point of the cam curve is large, the stability at the time of stop is poor, and the stop stabilization time has to be lengthened in practical use. The progress of the curve B from point p to point q will be discussed below. FIG.
In the calculation curve of the B curve, each curve of the acceleration change rate, the acceleration, and the speed is shown from the top. The horizontal axis is the time axis, and the elapsed time t of the stop point q is set to 0, and the time to the left of the q point is a negative number. Assuming that the time at an arbitrary point on the time axis is -t, the acceleration, velocity, and pulse number at -t are calculated.
In the numerical control device, the number of pulses corresponding to the length is controlled instead of the length, so that the unit of acceleration is pulse per second and the unit of speed is pulse per second.
Further, on the time axis, the stop point q is set to 0 and the left side is a negative number, so the point p is a negative time. The negative elapsed time from the stop point is -t seconds. If the acceleration is a pulse per second, the speed is v pulse per second, and the number of pulses from the stop point is S, the pulse number S is S ≦ 0, that is, a negative number. Also,
Since the acceleration change rate is a differential value of the acceleration, the acceleration change rate is δa / δt. Assuming that the acceleration change rate is a linear change and the change rate is −β, δa / δt = −β · t
(Equation 1) is obtained. Since the elapsed time is -t, the right side is a positive numerical value. From (Equation 1), δa = (− β · t) δ
t (Equation 2). Since the acceleration a is an integral value of the acceleration change rate, a = ∫δa + C. At the point of t = 0, since the acceleration is 0, C = 0. Substituting (Equation 2) and C = 0 into δa, a = ∫ (−β · t) δ
t. When this integral equation is calculated, a = −β · t ²
/ 2 (Equation 3). Since −t becomes positive when squared, the right side becomes a negative number. The negative number of the acceleration a is the deceleration. Since the speed v is an integral value of the acceleration, v = ∫a · δt +
C. At the point of t = 0, the speed is 0, so C = 0. Substituting the right side and C = 0 (Equation 3) in a, v = ∫ a ^{(-β · t 2/2)} δt. When calculating this integral equation is ^{v = -β · t 3/6} ( Equation 4). Since the cube of −t is a negative number, the calculation on −t is positive on the right side. Since the pulse number S is an integral value of the speed, S = ∫v · δt + C. At the point of t = 0, the number of pulses is 0, so C = 0. Substituting the right side of (Equation 4) and C = 0 into v, S = 、 (− β ·
a t ^3/6) δt. When calculating this integral equation, −
S = the β ^· t 4/24 (Equation 5). When the elapsed time t is calculated from the number of pulses S, the following equation is obtained from (Equation 5).
S) · 24 / β｝ Since the time t in the ^４ calculation is a negative time, taking −t gives −t = ｛(− S) · 24 /
β｝ ^1/4 (Equation 6). In (Equation 6), S>
When it is 0, the right side is an imaginary number and is not satisfied. (Equation 6)
Is set to t = − ｛(− S) · 24 / β｝ ^1/4 , and substituted into t in (Equation 4), v = −β [− ｛(− S) · 24
/ ^{Beta} 1/4]} Since 3/6 negative cubed is a negative number, v
= Β [{(- S) · 24 / β} 1/4] of 3/6.
To summarize the constants, v = {β (24 / β) ^3/4 / 6}
(−S) ^Assuming that the ^／ constant is ρ, Δβ (24 / β)
^3/4/6 ｝ = ρ, and (Equation 6) gives v = ρ · (−
S) ^3/4 (Equation 7). (Equation 7) does not hold when S> 0 because the right-hand side is an imaginary number and is not satisfied.
Is a negative number or 0. As the calculation method, since the calculation was performed with the stop point as the origin, the count number up to the stop point was a negative number. However, if the count number 4 of the deviation counter 3 in FIG. ) Is v
= Ρ · Z ^3/4 (Equation 8).

[0005]

[Operation] Since the rotation speed of the servo motor has a delay before reaching the rotation speed proportional to the speed command,
The curve when the speed command to the servomotor is set to v has a slight error with respect to the calculation formula. Therefore, it is necessary to set a speed curve with a small change in acceleration at the time of stop.
As a result of the experiment, by setting the speed command to the servo motor to the value calculated by (Equation 8), the stability at the time of stopping could be improved. Assuming that the residual count number until the stop is Z and an appropriate constant is ρ, the speed command v
Is examined with respect to a curve in which v = ρ · Z ⁿ (Equation 9). The curve A in FIG. 3 corresponds to the case where the index n = 1 because the count value of the deviation counter is directly applied to the servomotor via the DA converter in the case of the conventional numerical controller having the deviation counter. . The A-curve has a disadvantage that the stop stabilization time becomes redundant. The curve B in FIG. 3 is a case where the index n = 3/4. This is a curve in which the stop stabilization time is short and the stop stability is good. The curve C in FIG. 3 omits the progress of the integral calculation, but has an index n = ２. Since the acceleration change rate at the stop point does not become zero, the vehicle becomes unstable at the time of stop. To make the acceleration change rate at the stop point zero,
The index n must not be less than 2/3. The index n is 0.
When it is 7 or more, the acceleration change rate at the stop point becomes 0, and the stability at the time of stop can be improved. The index n = 1
In this case, since the stop stabilization time becomes redundant, the exponent n must be 0.9 or less in order to shorten the stop stabilization time. For the above reasons, when the index n is 0.7 to 0.9, the stop stabilization time can be shortened as compared with the conventional numerical control device having the deviation counter, and the stop time can be reduced. Good stability. The larger the constant ρ and the larger the index n, the longer the stop stabilization time, but the better the stability at the stop. FIG. 7 shows a speed waveform at the time of stop. Velocity waveform A
Shows an example in which the time to stop is long. Stop waveform B
Is an example of a suitable stop. A stop waveform C shows an example in which an overrun occurs at the time of stop and the operation stops unstable. The reason why the waveform becomes unstable at the time of stop is that after the servo motor inputs the speed command, the speed becomes the rotation speed corresponding to the speed command.
This is because there is a time delay. In order to shorten the stop stabilization time and improve the stability at the stop, the stop waveform B must be used. For that purpose, it is necessary to set the constant ρ or the index n in (Equation 9) to an appropriate value. By setting these to appropriate conditions, the stop stabilization time in the numerical control device having the deviation counter can be reduced.
It can be set to 0.02 seconds or less.

[0006]

[Embodiment of Calculation Formula] FIG. 5 shows an embodiment of the present invention. A function calculator constituted by a microcomputer or a signal processor is provided between the deviation counter 3 and the DA converter 5 in the block diagram shown in FIG. 12 are interposed. Reference numeral 13 denotes an output of the function calculator 12, which is a speed command value. The function calculator 12 receives the count value 4 of the deviation counter 3 and outputs the result of the function calculation as a speed command value 13. Assuming that the count value 4 of the deviation counter 3 in FIG. 5 is Z and the speed command value 13 of the function calculator 12 is V, the speed command value V and the rotational speed v (pulse per second) of the servo motor ignore the delay time. then because proportional, the proportionality constant and K, when the count value 4 of the deviation counter 3 and Z, from (equation 9), (a proviso n = 0.7~0.9) V = K · Z n ( Equation 10) can be derived. Proportionality constant K is a coefficient between the calculated value and the velocity command value V of Z ^n, is a proportionality constant which is set in advance. The calculation can be easily performed by calculating the speed command value V by (Equation 10) and outputting the calculation result to the DA converter as the speed command value 13.
FIG. 6 shows an embodiment using a microcomputer or a signal processor. In a numerical control device using a microcomputer or the like, it is common practice to perform similar operations using a function of a microcomputer or the like without separately providing circuits such as an oscillation circuit and a deviation counter. Reference numeral 14 in FIG. 6 denotes a microcomputer, which includes the oscillation circuit 1, the deviation counter 3, the function calculator 12, and the like in FIG.
The structure is replaced by the microcomputer 14. The microcomputer 14 performs a timer interrupt every predetermined time, for example, every 1 ms, and executes the following processing every time the timer interrupt occurs. A numerical value corresponding to the number of pulses for each fixed time in the oscillation circuit is added, and the total value is stored in the memory A. The generated pulses 11 of the rotary encoder 10 are counted and accumulated in the memory B. Since the value obtained by subtracting the cumulative value of the memory B from the cumulative value of the memory A corresponds to the count value of the deviation counter,
The value obtained by subtracting the cumulative value of the memory B from the cumulative value of the memory A is Z, and the result of calculating the speed command value V by (Equation 10) is output to the DA converter 5 as the speed command value 13. With any of the above means, in the numerical controller having the structure shown in FIG. 1 or a structure similar thereto, when the count value of the deviation counter is Z and an appropriately determined proportional constant is K, it is output to the DA converter. a speed command value V, the ^{V = K · Z n, n} = 0.7~0.9
It can be. In the embodiment, the speed command value for the servo driver is converted into a voltage and given. However, in the case of a servo driver in which the speed command is input by a digital signal or a pulse train frequency, the speed of the servo driver is According to the input method,
Method.

[0007]

[Embodiment for Stably Stopping in Short Time] FIG. 7 shows a waveform at the time of stoppage in the case where the present embodiment is implemented. The stop waveform A indicates a waveform in which the stop stabilization time is too long. The stop waveform B shows a state in which the stop stabilization time is short, the stop stability is good, and the stop is ideal. The stop waveform C indicates a state in which an overrun has occurred and an unstable stop has occurred. These waveforms are expressed by the speed command equation (Equation 1) to be adopted.
0) relates to the proportionality constant K and the index n. The greater the proportional constant K, the longer the stop stabilization time, but the better the stability at the stop. When it becomes smaller, the opposite effect occurs. Also, the larger the index n, the longer the stop stabilization time and the better the stability at the time of stop. The opposite effect occurs when the index n decreases. Therefore, the waveform at the time of stop is monitored, and the proportional constant K
Alternatively, the index n is appropriately changed. In this case, since it is difficult to change the two values, the influence of the proportionality constant K is large. Therefore, the constant n is changed and the proportionality constant K is changed. As for the method of monitoring the waveform, it is necessary to monitor the waveform up to the stop and the waveform after the stop. The time in the waveform can be monitored by the number of timer interrupts. Monitoring of the waveform up to the stop can be performed by the following means. The allowable time of the stop stabilization time is determined, and when the allowable time is exceeded, the proportional constant K is reduced. After the residual count value until the stop reaches a minimum value (for example, 5), the allowable time until the stop is determined. When the time exceeds this range, the proportional constant K is decreased. Monitoring of the waveform after the stop can be performed by the following means. During a minute time (for example, 5 ms) from when the count value of the deviation counter becomes 0, the proportional constant K is increased by the overrun count value.

[0008]

According to the numerical controller having the deviation counter,
As a means for controlling the movement of a tool of a machine tool or the movement length and rotation angle of a workpiece to stop at a fixed position, or a means for controlling the operation length and operation rotation angle of an industrial robot to stop at a fixed position It is widely and generally used. In these controls, the specified length or rotation angle must be controlled to stop at a fixed position, and then the next operation must be performed.Therefore, the shorter the stop stabilization time, the shorter the work time. Can be. In particular, when one operation is a high-frequency operation of 0.5 second or less, it is extremely important to shorten the stop stabilization time. In a conventional numerical controller having a deviation counter, a stop stabilization time usually requires 0.1 seconds, and the shortest one is
It could not be less than 0.05 seconds. The present invention
Since the stop stabilization time can be reduced to 0.02 seconds or less, in order to speed up the operation of a machine tool or an industrial robot using a numerical controller having a deviation counter,
The effect of the present invention is great.

[Brief description of the drawings]

FIG. 1 is a block diagram of a conventional numerical controller having a deviation counter. 1 is an oscillation circuit, 2 is an oscillation pulse output from the oscillation circuit, 3 is a deviation counter, and 4 is an output of the deviation counter, which is a count value. 5 is a DA converter, 6 is a voltage output of the DA converter, and is a speed command. 7, a servo driver; 8, a motor; and 9, a power. 10 is a rotary encoder. Reference numeral 11 denotes a pulse generated by the rotary encoder.

FIG. 2 is a graph in which a horizontal axis is a time axis, and a pulse per second corresponding to a speed is shown on a vertical axis.
, And the dotted line indicates the pulse per second of the generated pulse 11 of the rotary encoder. p indicates the stop point of the oscillation pulse 2 and q indicates the stop point of the output pulse of the rotary encoder. The time from point p to point q is the stop stabilization time.

FIG. 3 is a curve of an acceleration change rate, an acceleration, and a speed, with the horizontal axis as a time axis, for a curve A, a curve B, a curve C, and a curve D, and a curve A is a curve of a conventional numerical controller having a deviation counter It is. Curve B is a curve adopted by the present invention, and curve C is a curve having a large acceleration change rate at a stop point. A curve D indicates a deceleration curve of a deformed sine curve or a deformed trapezoidal curve which is a general cam curve.

FIG. 4 shows curves of acceleration change rate, acceleration, and speed, with the horizontal axis as a time axis, for the curve B in FIG. 3 to be adopted by the present invention. The acceleration change rate, acceleration, speed, pulse number, and the like are obtained for an arbitrary point -t on the time axis.

5 is a block diagram of one embodiment of the present invention, in which a function calculator 12 and a speed command value 13 are inserted between the count value output 4 of the deviation counter of FIG.

FIG. 6 is a block diagram of a microcomputer-based embodiment of the present invention.
The function calculator 12 is replaced by 14 microcomputers.

FIG. 7 shows various stop waveforms with the horizontal axis as the time axis and the vertical axis as the speed. The stop waveform A shows a stop waveform when the stop stabilization time is too long. The stop waveform B has a short stop stabilization time and shows a target stop waveform. The stop waveform C shows an unstable stop waveform that overruns due to sudden stop.

Claims (2)

[Claims] In a numerical control device having a deviation counter, in a numerical control device having a structure shown in FIG. 1 or a structure similar thereto, the count value of the deviation counter is Z and an appropriately determined proportional constant is K. At this time, the speed command value V given to the servomotor via the DA converter is expressed as V = K ·
And Z ^n, the structure of the numerical controller being characterized in that the index n = 0.7 to 0.9. 2. The numerical controller according to claim 1, wherein the proportional constant K in the equation of claim 1 is changed so as to monitor the waveform at the time of stop and stably stop in a short time.