JPS59109864A - Slow speed integration-type speed detector - Google Patents

Abstract

PURPOSE:To make such control possible that a speed integral value is equal to the value of the position by outputting a speed so that the speed integral value does not exceed a position moving length in the low-speed region where no position detection pulse is inputted within a certain time. CONSTITUTION:In accordance with the counted value of position pulses in a certain period due to a counter 60 and the counted value of intervals of position pulses due to a counter 62, a CPU64 calculates a position moving length P and a speed V in every certain period and calculates a speed integral value S also. In the low-speed region where no position pulses are detected by the counter 60, the CPU64 outputs the preceding speed V stored in a memory and defines V+S as a current integral value in case of P>=S+V. If the counted value of a counter 61 is zero in the next period also, the similar processing is performed; and when P<S+V becomes true, the CPU64 outputs P-S=V as a speed and defines the integral value S as the length P. Consequently, the ingegral value S is controlled so as to be equal to the length P, and integral elements and operation error are absorbed to perform the accurate speed control, and a servo motor is turned smoothly without oscillation.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a speed detection device that detects speed information from a position detection pulse generated every unit movement obtained from a position detector.

The speed is detected from the position detection pulse period, and the
(For servos and mechanisms that perform speed control by applying speed feedback, it is necessary to detect speed quickly and accurately, but in the low speed range, accurate speed detection and control is difficult, and it is difficult to smooth the movement of the nervo motor.) This was not possible.This is because in a servo mechanism, when integrating the speed, it should match the moving position, but due to errors due to calculation accuracy and errors caused by the integral element of the speed control section, the returned speed may not match. This is because the integral value of the information does not match the position, which causes vibration.

SUMMARY OF THE INVENTION Therefore, the present invention provides a speed detection device that outputs a controlled value such that the estimated speed value becomes equal to the position value so as to smooth the movement of the servo motor in the low speed range. be.

The present invention uses the number of position detection pulses input within a certain period to estimate the position movement distance during that period, and while the position detection pulses are input within a certain period, the position detection pulse cycle If the position detection pulse is not input within a certain period of time, the last calculated speed will be output. Only when the speed is smaller than the position movement distance during the period in which the position detection pulse was last input, the velocity displayed in the first image is output, and when the integral value becomes large, the position movement distance and the previous position movement distance are output. This function outputs the difference between the integral value of This is the speed output.

Hereinafter, the present invention will be explained in detail based on examples.

FIG. 1 is a principle diagram of the first method for obtaining J5 speed information in the present invention, where P is a reference pulse whose time unit is the basic period To, and PP is a position detection pulse. .

In this first method,! ;! The position detection pulse interval immediately before the generation of the Qt pulse TP is measured, and the value is calculated at the reference pulse generation point (period 1 of the position detection pulse {JT). , at the time when the reference pulse TP2 is generated, the value of T3 is equal to the period T at that time.
Then, at the time when TPa occurs, the value of Tθ is the period T at that time. Speed information (2) is obtained from the reciprocal of the period T thus obtained. That is, V=K+
/T......(1) (K1: constant of proportionality).

FIG. 2 shows a second method for obtaining speed information, in which the number N of position detection pulses PP input during a reference period To in units of time is counted. Also, the interval T1 between the position detection pulse S1 immediately before the period and the reference pulse generation point (the starting point of the period) is visually measured. Furthermore, the interval T2 between the last position detection pulse S4 of the cycle and the end point of the cycle 1yJ is measured.

Then, from these measured values, the time T required for the number N of position detection pulses to be input is determined using the following formula.

1-=1-o+T1-'r2・・・・・・・・・(2)
Dividing this time 1F by the number of pulses N gives the period per position detection pulse, and the reciprocal of the period becomes the frequency of the position detection pulse, that is, speed information.

Therefore, if the speed is different from V, 'a degree V is V=K1
・N/T・・・・・・It is determined by (3) (K+ is a constant of proportionality).

In this manner, when the position detection pulse is input within the reference period 1.0, the speed information shown in equation (1) or (2) above is calculated and output from the speed detection device. However, when the position detection pulse is not input within the reference period To, the present invention calculates the speed of the reference period in which the position detection pulse was last input, so that the integral value of j* degrees is the reference period + 1I.
If it is smaller than the position movement distance within J, it is output, and if it is larger, the difference between the position movement distance and the integral value is output.

However, if the position detection pulse is input within the base semi-period 0, the velocity V is calculated as shown in equations (1) and <3), and at the same time the number of position detection pulses is Multiply N by the proportionality constant 1<2, P=K2・N...
- According to (4), the position movement distance between the reference circumferences To is determined and stored in the memory. moreover,
Calculate the integral of speed information + tfIS, position detection pulse 1]
When one or more P is detected during the reference period TO, the second integral value S is set to "0".

For example, in FIGS. 1 and 2, time point -1-
Assuming that the position detection pulse PP is not input during the reference period TO from P4 to the time TPt, the position detection pulse was input] - P4 time (N4≠ON5-0...
...Nt=O) is output, and the integral value S is obtained by adding the velocity V for each reference pulse. That is, Vt=V4St=St-++V4・・・・・・・・・(
5).

When the integral value St obtained in this way becomes larger than the previously stored position movement distance P4, the speed information Vt is as follows: <St-Sz-++V)>(P4=K2N4) The difference Vt between the position movement distance tlllltP4 and the tree branch value Si-+ from time t before the Lee reference period is calculated as follows: Vt=P4-St-+.

As a result, the minute value SL is St=P4.

In this way, in order not to exceed the positional movement distance mp, rounding is performed so that the integral value of the speed matches the positional movement distance.

As described above, the present invention is applicable to the point in time when the position detection pulse is manually applied during the previous reference period 1-θ (in FIG. , time point) position movement distance #[P=K2N
The value of is controlled so that the integral value of the velocity during the period in which no position detection pulse is input thereafter does not exceed.

FIG. 3 is a block diagram of a first embodiment of the present invention that controls and outputs smallpox information based on the first method.

In FIG. 3, 60 is an up-down counter, which counts up the positive direction rotation position pulse PP from the position detector by 1.
... and count the reverse rotation position pulse PP'
61 is the A register, and the counter 1 value of the up counter 60 is set to 1 by the reference pulse P. 62 is a timer counter that counts clock pulses CP. 63 is a B register that counts position pulses PP. The count value of the timer counter 62 is set in the B register 63 by PP', and then the timer counter 62 is cleared.

Reference numeral 64 denotes a performance device consisting of a microprocessor, which reads the values of the A register 61 and the B register 63 for each reference pulse Tl) whose time unit is the reference cycle]1lJTo, and outputs speed information.

65 is a C register, which temporarily stores the output speed information of the arithmetic unit 64, and 66 is a register for digital/analog (O
A) A converter, which converts the contents of the C register 65 into an analog value. 67 is or game 1~.

In the above J; C configuration, the position detection pulse PP is
When the 1% Lla pulse TP is input to the counter 60 and counted, the value of the counter 60 is changed to the A register 61.
After that, the counter 60 is cleared and the remelting position detection pulse PP is decremented. As a result, as shown in the principle explanatory diagram of FIG. -3, there are four position detection pulses between reference pulse TP+ and TP2.
The register 61 stores the number 4 of the position detection pulses as the reference pulse TP2. Similarly, reference pulse 1-P3
At the time of occurrence, the A register 61 stores 3). Therefore, by reading the stored value N of the A register 61 for each reference pulse TP and multiplying this stored value N by the proportionality constant K2 as shown in equation (4) above, the position within this base i1ji period To can be calculated. It is possible to obtain the moving distance P. Further, the timer counter 62 increments the clock pulse CP by tI and is cleared by the position pulse PP applied via the Δagut 67. Also, the count of the timer counter 62 {white is
Set 1 to B register 63 by position detection pulse PP
Therefore, the B register 63, as shown in FIG.
The position detection pulse interval immediately before the reference pulse is generated is stored. In other words, the base L((The value 1-3 is stored for the pulse TP2 generation line, and the value -1-e is stored for the T1 and 3 generation line.

Since this position detection pulse interval is considered to be the period T of the position detection pulse at the time when the reference pulse is generated, the velocity information V can be obtained by taking the reciprocal of this period.

Then, the present invention integrates this speed information V and outputs speed I odor information V by controlling the integrated value S so that it does not exceed the above-mentioned position movement distance P.

Next, the operation of applying the configuration of the embodiment shown in FIG. 3 will be explained along with the operation flow.

4(a) and 4(b) show the flow of the first embodiment of the present invention, in which the arithmetic unit 64 calculates the number N of position detection pulses set in the A register 61 every time the reference pulse TP is generated. It also reads and stores the counter value of the register 63, ie, the position detection pulse T.

Next, it is determined whether the number N of position detection pulses read earlier is "0" or not. K1/]- to calculate the speed information V, and change the speed information V at the time of the previous reference pulse to the speed information obtained this time as the detected speed information VO, and set Vo=V.

Next, a proportional constant K is added to the position detection pulse number N read earlier.
Multiplying by Change it to distance P and memorize it. Further, the integral value S is set to "○" and is stored in the memory.

The speed information Vo thus obtained is sent to the C register 65, converted into an analog value by the DA converter 66, and sent out as speed information.

In this way, the position detection pulse PP is input to the up/down counter 60, and the number of position detection pulses is read from the △ register 61. , position movement distance 1!lltP.The integral value S is calculated, rewritten, and stored in the memory!
At the same time, speed information Vo is output from the DA converter 66.

Next, the position detection pulse PP is not input to the up-down counter 60 within the reference period To, and the up-down counter 60
The reading value of the displacement detection pulse number from the J counter 60 is 1.
0'', the position movement distance P during the reference cycle period To stored in the memory by the previous reference pulse and the integral value S in the memory (note that the integral value is calculated by inputting the position detection pulse PP) Since the maximum speed is rOJ, the beginning is rOJ.> is compared with the last speed information V previously stored in the memory.

If the positional movement distance P is too large, then I〕≧S-1
-V, the speed information outputs the last speed information V previously stored in the memory, and the integral value S is the sum of the speed information V added to the previously stored integral value S. Change to That is, let the memory contents of the integral value be s=V+S. and,
Even in the next reference pulse, the number of reading position detection pulses N is r
OJ, and if the sum of the newly updated integral value S in the memory and the previously stored speed information V is smaller than the already stored position quotient movement distance P, then the above-mentioned As such, the last speed information V stored is output, and the integral value S is stored by adding the speed information V to the previous partial value S. Then, the value of S-1-V is greater than the positional movement distance P. Then, the integral value S of the speed is subtracted from the position movement distance P, the value is output as speed information V, and the integral value S of speed 1 is set as the position movement distance FIAP. In other words, the fact that p-s=v means that the value S of the speed increases by the unit of speed information V stored at the time of the reference pulse generation, so the value of P-S becomes 1 of this speed information V. If V of 1 unit is added to the integral value, the integral value S of the velocity becomes large and the velocity exceeds the position movement distance P. The value obtained by subtracting the integral value S is output as the detected speed, so that the integral value of the detected speed always does not exceed the position movement distance.

FIG. 5 is a block diagram of an embodiment of the present invention when using a second technique for determining speed information. This example and the third
The difference in the configuration of the embodiment shown in the figure is that the B register 63 is set to
What is the position detection pulse in the configuration of the embodiment shown in FIG. 3 has been changed to the reference pulse TP.As a result, what is stored in the B register 63 is:
T1 or T2, which is not shown in FIG. 2, is stored. That is, the timer counter 62 is cleared every time the position detection pulse PP is input, and the timer counter 62 is cleared every time the position detection pulse PP is input, and the timer counter 62 is cleared every time the position detection pulse PP is input.
are being counted. Therefore, for example, counting starts from position detection pulse S4 in FIG.
Accordingly, the value 2 of the timer counter 62 is let into the B register.

As a result, the calculation device 64 of this embodiment performs the process shown in equation (2), calculates the period, and calculates the cycle using equation (3) from the number N of position detection pulses determined by the A register. The speed information is obtained by performing the processing shown in FIG. Therefore,
The processing flow in the arithmetic unit 64 also includes the process of calculating the above-mentioned T, and the flow in FIG. 4(A) changes to FIG. 6.・Other than changing it to N/Dou, it will be left untouched.

Each time the reference pulse TP is generated, the arithmetic unit 64 registers the register 6.
The number N of position detection pulses set to 1 is read, the count value T2 of the register 63 is read, and the calculation shown in equation (2) is performed.

That is, the previous reference pulse T. P and add the previous count value of the register 63, 1, which was stored in the memory in the arithmetic unit 64,
The required input time T is determined by subtracting the count value T2 of the register 63 read with the current reference pulse.

Next, T1 stored in the memory is changed to T2.

Hereinafter, after the point ■, the process is the same as in FIG. 4 (opening) except that ■ is changed to V=K+·N/T.

In addition, in a3 in the low speed range, assuming that the number of position detection pulses input during the reference period To is one or less, and assuming that the position movement distance P = K2 · 1 - K2, as shown in Figure 4 (■) ,
There is no need to calculate and store the position movement distance for each reference pulse, and when the number of position detection pulses N reaches rOJ, P≧
J: Yes, if we make the judgment that K2≧S0V instead of the judgment that S+V.

As described above, the present invention outputs the speed 1 information so that the integral value of the speed does not exceed the position information, and
In the end, we made the integral value of the speed equal to the position value, so we absorbed the error caused by the integral element of the control unit and the error caused by the calculation accuracy, and performed accurate speed control. →
Novo Motor rotates smoothly without causing any vibration.

[BRIEF DESCRIPTION OF THE DRAWINGS] FIG. 1 is a principle diagram of the first method for obtaining speed information according to the present invention, FIG. 2 is a principle diagram of the second method, and FIG. 3 is a diagram of the principle of the second method according to the present invention. A block diagram of the first embodiment of the present invention based on the first method, FIG. FIG. 6 is a block diagram of the second embodiment of the present invention when using the same, and is an operation flow of the third embodiment. "P...M semi-pass/L/pass, PP...position detection pulse, CP...clock panorez, 6o...uptown counter, 61...A register, 62...timer counter, 63 ...B register, 64... Arithmetic device, 65... C register, 66... D/A converter. 1364-13651

Claims (4)

[Claims] (1) A speed calculation means for extracting speed information from position detection pulses generated during unit movement ffl, a speed integration means for integrating the speed information, a means for determining whether or not the
When the position detection pulse is inputted by the discrimination means, the speed information from the stray detection means is outputted, and the speed integral value of the above-mentioned)*degree integration means is calculated by the clino 7, and the position detection pulse is If not input, the speed information in the reference period in which the position detection pulse was inputted to Moe from then on will be used as the corresponding 1;! : quasi-periodic) J1 degree information, and divide the velocity information by the velocity integrator Ω, and if the fraction {10} is smaller than a certain set value, output the velocity information and set If the value exceeds the set value, a speed integral value in a reference cycle 1111 immediately before the reference cycle is subtracted from the set value, and the lc value is outputted as speed information. (2) The set value is a constant value added to the number of input position detection pulses in a reference period immediately before a reference period in which no position detection pulses are input, and is equal to 1C. The speed detection device described. (3) The speed detection device according to claim 1 or 2, wherein the speed calculation means measures the position detection pulse period in gJ and outputs speed information based on the reciprocal of the period. (4) The speed calculating means measures the number of position detection pulses input within a certain reference period and the required input interval required for inputting the number of position detection pulses, and calculates the number of position detection pulses according to the required human power. A speed detection neck device according to claim 1 or 2, which outputs speed information by dividing by time.