JPH05341847A - Function generating device for positioning control - Google Patents

Abstract

PURPOSE:To calculate the acceleration/deceleration functions at a high speed and with high accuracy and also at the comparatively low cost without using such an expensive arithmetic-only processor such as a DSP. CONSTITUTION:In regard of a function generator which generates the functions of the linear acceleration/deceleration and the S-shaped acceleration/deceleration for the positioning control, a CPU 1 generates a positioning control pattern with a shift average operation of a single step or plural steps. Furthermore the CPU 1 can generate a high accuracy command through the fraction processing carried out in the shift average operation state. The ring buffer memories 3 and 4 are used for the shift average operation carried out in plural steps. That is, a linear acceleration/deceleration pattern is obtained with the shift average operation of the first step with an S-shaped pattern obtained with the shift average operation of the second and subsequent steps respectively to a stepped input pattern. In such a way, the S-shaped acceleration/deceleration control is attained with no shock.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a function generator for positioning control for controlling the rotation speed of a motor and acceleration / deceleration control of the rotation speed, and more particularly for positioning control for performing linear or S-shaped acceleration / deceleration control. Function generator

[0002]

2. Description of the Related Art A general example of a conventional positioning device for performing acceleration / deceleration control is shown in FIG. This example is a positioning system in which the slave shaft motor 10 is rotated at a rate synchronous with the rotation speed of the spindle motor 5. When the spindle motor 5 rotates, the incremental encoder 6 outputs a pulse proportional to the amount of rotation. This pulse is converted into a voltage by an F / V (frequency-voltage) converter 21, and this voltage causes C-R.
The driven shaft motor 10 is rotated via the filter 22, the amplifier 23, and the current amplifier 9. The acceleration / deceleration control is performed by the time constant of the CR filter 22. SW is a switch for switching “start / stop” of the slave shaft motor 10, and is connected to the F / V converter 21.

FIG. 2 shows another example of the conventional positioning device proposed in JP-A-2-121005 and JP-A-2-121006. In this example, the high-order computer 41 is a general-purpose microprocessor, gives a command command to the low-order computer 42, and the low-order computer 42 executes the command command. The lower-order computer is a DSP (Digital Signal)
This DSP is capable of performing 16-bit fixed point multiplication or multiply-accumulate calculation in 100 ns to 200 ns, is 5 to 10 times faster than a general-purpose microprocessor, and has a DC motor 47 The rotation angle is fetched from the counter 45 using the encoder 46, servo calculation is performed inside the DSP, the manipulated variable is output to the D / A converter 50, and the DC motor 47 is rotationally driven via the current amplifier 51. That is, the low-order computer 42 obtains the operation amount based on the digital servo control system constituted by software based on the command given from the high-order computer 41 and the current position information from the counter 45. As described above, in this conventional configuration, the low-order computer 42 uses the DSP for performing high-speed calculation, and performs the function calculation for positioning having the S-curve acceleration / deceleration function.

[0004]

FIG. 3 shows a speed command voltage pattern in the conventional configuration of FIG. 1 described above. Spindle motor 5
Is rotated at a constant speed, and the slave shaft motor 10 is intermittently started. 3A shows the output pattern from the F / V converter 21, and FIG. 3B shows the waveform of the output pattern from the CR filter 22. As shown in the figure, since the acceleration / deceleration control by the CR filter 22 is an exponential acceleration / deceleration, there is a shock at the start of acceleration and at the start of stop,
Further, there is a problem that the acceleration / deceleration time becomes longer when the time constant is increased.

On the other hand, the conventional example shown in FIG. 2 uses a dedicated processor such as a DSP, and therefore has a problem that the apparatus becomes extremely expensive.

In view of the above points, an object of the present invention is to provide a positioning control function generator capable of performing an acceleration / deceleration function operation with high accuracy and at a relatively low cost.

[0007]

In order to achieve the above object, the present invention is a positioning control function generator for performing motor rotation speed control and speed acceleration / deceleration control. Alternatively, it is characterized by including a moving average calculation means for performing a moving average calculation in one step or a plurality of steps for converting into an S-shaped acceleration / deceleration pattern.

As one form thereof, the present invention is characterized by further comprising a fraction processing unit for performing a fraction process during the moving average computation of the moving average computing unit.

[0009]

According to the present invention, the stepwise speed command calculation value is subjected to the function calculation for the positioning for performing the speed control and the acceleration / deceleration control by the moving average calculation means of one stage or a plurality of stages, and the linear acceleration / deceleration and A function (control pattern) for S-curve acceleration / deceleration is generated.

In the moving average calculation, the number of buffers in the ring buffer memory for storing sampling data is n.
If it is (integer), it is calculated by the following formula.

[0011]

[Equation 1] (moving average) = {(added value up to the previous time)-(oldest sampling data) + (current sampling data)} / n When one-step moving average calculation is performed on the step-like input pattern The linear acceleration / deceleration pattern has an S-shaped pattern in the moving average of the second stage and above.

That is, assuming that the number of buffers in the ring buffer memory is n1, n2, ..., Ni, the output change value Vi at the final stage with respect to the change V0 of the step-like command at the first stage.
Is

[0013]

## EQU2 ## Vi = V0 × (1 / n1) × (1 / n2) × ...
Since it can be obtained by × (1 / ni), shockless acceleration / deceleration control by S-shape can be performed even when the command value of the first stage is changed stepwise during the operation like the override performed by the numerical control.

As described above, since the moving average calculation according to the present invention can be executed by a simple equation and it is not necessary to solve a complicated equation, the S-shaped pattern of the S-shaped pattern can be obtained without using an expensive arithmetic-dedicated processor such as a DSP. Such acceleration / deceleration function calculation can be performed at high speed.

Further, by performing a singular calculation during the moving average calculation of each stage, it is possible to perform a function calculation for acceleration / deceleration control that gives a highly accurate position command.

[0016]

Embodiments of the present invention will now be described in detail with reference to the drawings.

(First Embodiment) FIG. 4 shows the configuration of an embodiment of the apparatus of the present invention in an open loop system. The first embodiment is a positioning system for rotating the slave shaft motor 10 at a ratio synchronous speed with respect to the main shaft motor 5 rotating at a constant speed. The driven shaft motor 10 performs intermittent operation according to an external command, but is actually applied to a fixed-length cutting device or the like that cuts an object to be cut (not shown) sent to a predetermined position into a predetermined length.

The microcomputer 1 has a CPU (central processing unit) 11, a ROM 12 for a program and a RAM 13 for a work area, and an incremental encoder 6 is provided for each CLK (clock) signal input from the sampling pulse generator 2. The amount of rotation of the spindle motor 5 at the sampling time t is measured by reading the value of the pulse counter 7 that counts the pulses from. The microcomputer 1 calculates the rotation speed vm of the spindle motor 5 from the next sampled measurement data, and the spindle motor 10
D / A setting for rotating the
This is performed for the (digital / analog) converter 8. The D / A converter 8 converts the setting data from the microcomputer 1 into an analog value, and via the current amplifier 9, the slave shaft motor 1
Rotate 0 at the specified speed. SW is the slave shaft motor 1
It is a control switch for controlling "start / stop" of 0.

In this embodiment, the microcomputer 1 of the present embodiment has a pair of ring buffer memories 3 and 4.
Is used to perform the moving average calculation process for the speed command in two stages as described below.

That is, in the device of this embodiment, the rotation speed vm of the main shaft motor 5 is set to the rotation speed v of the slave shaft motor 10.
Rotate with s. Note that the relationship between vm and vs is expressed by the following equation (1) when the proportional constant is k1.

[0021]

[Mathematical formula-see original document] vs = k1 + vm (1) The rotation speed vm of the spindle motor 5 is given by the following equation (2) by the pulse number p from the incremental encoder 6 at the sampling time t, where the proportional constant is k2. Required to.

[0022]

Vm = k2 × (p ÷ t) (2) The proportional constant k2 is obtained from the number of output pulses of the incremental encoder 6 per one revolution of the spindle motor 5, or the like.

Next, the transition of the speed command data to the slave shaft motor 10 when the switch SW for starting control of the slave shaft motor 10 is switched from OFF (open) to ON (closed) to OFF (open). This will be described with reference to FIG.

First, it is assumed that the switch SW is OFF and the data in the two ring buffers 3 and 4 are all "0". The microcomputer 1 is
If the input signal from the switch SW (hereinafter referred to as SW input) is OFF, the pulse counter 7 at the unit time t
Is forcibly set to "0". That is, in the following calculation, the rotation speed vm of the spindle motor 5 is set to "0".

If the spindle motor 5 is rotating at a constant speed vm, the microcomputer 1 detects the increment of the pulse counter 7 only while the SW input is ON. It is detected as shown in (1). When the first-stage moving average calculation by the first ring buffer memory 3 is performed according to the procedure of FIG. 6 described later, the rotation speed vm ′ of the spindle motor 5 is linear acceleration / deceleration as shown in (2) of FIG. Become. Further, when the moving average calculation of the second stage by the second ring buffer memory 4 is performed according to the procedure of FIG. 6 described later, the rotation speed vm ″ of the spindle motor 5 is as shown in FIG.
As shown in (3), the acceleration / deceleration is performed in a curved shape (referred to as an S-shape). The speed command to the slave shaft motor 10 is vm ″
Since the calculation of the data of (1) is performed on the data of (1) and the data is output, the rotation of the slave shaft motor 10 has a waveform corresponding to (3) of FIG. 5, and shockless acceleration / deceleration can be realized.

Next, the procedure of the moving average calculation according to the present invention using a plurality of ring buffer memories will be described with reference to the flowchart of FIG. Note that P in P1, P2 ... Means a program step. This calculation is performed every time the CLK signal from the sampling pulse generator 2 is input. First, when the sample interrupt is started at P1, the ring buffer pointer is read at P2. This ring buffer pointer indicates the storage address of the oldest sampling data. At P3, the oldest data is subtracted from the total value of the data in the ring buffer memory, and the next P
In 4, the sampling data of this time is added to the result.
At P5, the total value is divided by the number n of ring buffers to obtain an average value, and this average value is calculated as the rotation speed vm '(for the first stage) or vm "(for the second stage) of the spindle motor 5. That is, the arithmetic expression is the following expression (3),
It becomes like (4).

[0027]

EQUATION 5 SUMi = SUM _i-1 −OLD + NEW (3)

[0028]

Vm ″ = SUMi ÷ n (4) SUMi is the current value of the sum of the ring buffers, SUM _i−1
Is the total value of the ring buffer up to the previous time, OLD is the oldest sampling data, and NEW is the current sampling data.

At the next P6, the oldest sampling data OLD is rewritten with the current sampling data value NEW, and at P7, the ring buffer pointer is counted up. In P8, it is checked whether or not the value of the ring buffer pointer exceeds the number n of buffers in the ring buffer. If it exceeds, the buffer start address is set in P9 in P9. P if not exceeded
Fly from 8 to P10. Although the ring buffer pointer is written in P10, the pointer at this point indicates the pointer of the oldest data in the ring buffer. At P11, this interrupt processing ends.

FIG. 7 shows the calculation result with actual numerical values. In this example, the number of buffers n of the ring buffer memories 3 and 4 is 10 in both the first and second stages.

(Second Embodiment) FIG. 8 shows the configuration of an embodiment of the apparatus of the present invention in a positioning system. This second
In the configuration of the embodiment, the motor 10 is set by the data setter 31.
The movement amount data and the movement speed data are set in the microcomputer 1 as numerical data. The microcomputer 1 includes a plurality of n ring buffer memories 34-3.
6, and the same number n of single data storage memories 37
Based on the above numerical data, the unit time t
Calculate the command pulse increment value in. This calculation is performed every time the CLK signal is input from the sampling pulse generator 2, and the calculation result is output to the deviation counter 32 as a command value every unit time t. The deviation counter 32 outputs the command position data based on the command pulse increment value (command value) and the motor 1 based on the feedback pulse from the incremental encoder 33 that outputs a pulse proportional to the rotation amount of the motor 10.
The value based on the difference between the current position data of 0 and the D / A converter 8
Output to. This output value is converted into an analog value by the D / A converter 8, and further the motor 10 is rotated via the current amplifier 9. The rotation speed of the motor 10 is determined by the command pulse increment value per unit time t.

The acceleration / deceleration control operation of the motor 10 in the apparatus of this embodiment will be described with reference to FIG. First, the microcomputer 1 converts the movement amount data into the pulse amount P,
The moving speed data is converted into the frequency f. In the stepwise command, the command pulse amount p at the unit time t is obtained by the following equation (5).

[0033]

## EQU00007 ## p = f.times.t (5) The microcomputer 1 first outputs a stepwise command pulse increment value p for each unit time t as shown in (1) of FIG. The final command pulse amount pe is the remainder obtained by dividing the moving pulse amount p by the pulse increment p in unit time.

This stepwise command data (p) is used as the first
Ring buffer 34 and fraction data storage memory 37
When the moving average calculation of the first stage is performed according to the procedure shown in FIG. 10 described below, the linear acceleration / deceleration pattern shown in (2) of FIG. 9 is obtained. Further, when the moving average calculation of the second stage is performed by the ring buffer 35 and the fraction data storage memory 38 of the second means according to the procedure of FIG. 10 described later, the S-shaped acceleration / deceleration pattern shown in (3) of FIG. Becomes After that, n
The output curve becomes smoother every time the moving average is calculated up to the stage, and a shockless acceleration / deceleration pattern with less change in acceleration is obtained.

By the way, in the positioning of this embodiment, the total value of the command pulse amounts to the deviation counter 32 needs to be accurately managed. The procedure of the moving average calculation according to the present invention including the fraction calculation processing for that purpose will be described with reference to FIG. Note that P in P21, P22, P23 ... Means a program step. First, in P22 and P23, the average value (SUM / n) is obtained from the total value (SUM) of the ring buffer memories, as in the processing of P2 to P5 in FIG. 6 described above. In P24, the remainder (fraction) in the average calculation is added to the fraction data up to the previous time (that is, the added value of the fraction data). In P25, the total value of the fraction data is the number of buffers n
If above, add 1 to the average calculated value (output value) at P26,
Further, n is subtracted from the total value of the fraction data. If the number of buffers is n or less, jump to P27. The subsequent processing of P27 to P31 is the same as that of P6 to 10 in FIG.

FIG. 11 shows a calculation result by actual numerical values based on the procedure of FIG. Here, the first step input value is the stepped command pulse amount, the sum value of the ring buffer memory is SUM, the fractional numerical value data after the fractional value arithmetic processing is used for the fractional value, and the moving average operation including the fractional value processing is performed for the output value. Shows the output value of. The number of buffers in the ring buffer memory is 10 in both the first and second stages.

[0037]

As described above, according to the present invention,
The following effects can be obtained.

(1) The basic processing of the moving average calculation is (3)
Since the equations and (4) are simple and there is no need to solve complicated equations, the acceleration / deceleration function calculation can be performed at high speed without using a dedicated processor such as a DSP.

(2) If the number of buffers in the ring buffer memory is n1, n2, ..., Ni, the output change value Vi at the final stage with respect to the change V0 of the step-like command at the first stage is

[0040]

Equation 8 Vi = V0 × (1 / n1) × (1 / n2) × ...
Since it can be obtained by × (1 / ni), shockless acceleration / deceleration control by S-shape can be performed even when the command value of the first stage is changed stepwise during the operation like the override performed by the numerical control.

(3) Further, when calculating the moving average of each stage,
By performing a fractional calculation, a function calculation for acceleration / deceleration control that gives a highly accurate position command can be performed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of a conventional positioning device.

FIG. 2 is a block diagram showing another configuration example of a conventional positioning device.

FIG. 3 is a waveform diagram showing a speed command voltage pattern in the conventional example of FIG.

FIG. 4 is a block diagram showing a configuration of an exemplary embodiment of the present invention.

5 is a waveform diagram showing a transition of speed command data to the slave shaft motor of FIG.

FIG. 6 is a flowchart showing a procedure of a moving average calculation according to an embodiment of the present invention.

FIG. 7 is a diagram showing a calculation result with actual numerical values according to the procedure of FIG.

FIG. 8 is a block diagram showing the configuration of another embodiment of the present invention.

9 is a waveform diagram showing a transition of output according to the embodiment of FIG.

FIG. 10 is a flow chart showing a procedure of a moving average calculation including a fraction calculation process of another embodiment of the present invention.

11 is a diagram showing a calculation result with actual numerical values according to the procedure of FIG.

[Explanation of symbols]

 1 Microcomputer 2 Sampling pulse generator 3 Ring buffer memory 1 4 Ring buffer memory 2 5 Spindle motor 6 Incremental encoder 7 Pulse counter 8 D / A converter 9 Current amplifier 10 Slave shaft motor 11 CPU 12 ROM 13 RAM 31 Data setter 32 deviation counter 33 incremental encoder 34 to 36 ring buffer memory 37 to 39 fraction data storage memory

Claims (2)

[Claims] 1. A positioning control function generator for performing motor rotation speed control and speed acceleration / deceleration control, wherein one step for converting a stepped speed command pattern into a linear or S-shaped acceleration / deceleration pattern. Also, a positioning control function generating device is provided with a moving average calculation means for performing a moving average calculation in a plurality of stages. 2. The function generating device for positioning control according to claim 1, further comprising a fractional processing arithmetic means for performing a fractional processing at the time of the moving average arithmetic operation of said moving average arithmetic means.