JP2000227437A - Rotation speed detector - Google Patents

Abstract

(57) [Problem] To provide a rotation speed detection that can easily obtain a high-accuracy and good response speed detection value, and that can be applied to a motor speed control device to obtain stable control. . SOLUTION: A limit value calculator 8 which adds 1 to a pulse difference value ΔP detected for each sampling period and divides the value by a sampling period Ts to output as a limit value Nlim, and uses a limit value calculator 8 for detecting a rotation speed. The rotational speed is directly obtained by dividing the difference value ΔP of the pulse supplied from the rotary encoder attached to the shaft for each sampling period and the divided value Nd of the difference value ΔT of the pulse input time supplied from the time counter 2 and the latch circuit 3. , The rotation speed is processed as the limit value Nlim by the limit 9 and then output as the rotation speed detection value Ndet.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a rotation speed detection device using a rotary encoder, and more particularly to a rotation speed detection device suitable for a motor speed control system using a vector control method.

[0002]

2. Description of the Related Art In various automatic control systems such as an industrial robot, a motor control device for performing feedback control according to a rotation speed command externally used is widely used.

FIG. 2 shows an example of a speed control device 20 in such an electric servo control system. As shown in FIG. 2, a speed detecting sensor is provided on a rotating shaft of a motor (electric motor) 23 to be controlled. 24 is directly attached and its output is taken into the speed control device 20 so that the feedback control is performed so that the rotation speed of the motor 23 converges to the rotation speed given by the speed command value Nref supplied from the outside. Has become.

The sensor 2 incorporated in the speed control device 20
4 is output from the speed detector 25 by the degree detector 25 first.
And supplied to the comparison input of the comparator 26, where the deviation from the speed command value Nref is obtained, and the deviation is supplied to the speed controller 21.

Therefore, the speed controller 21 is provided with a current amplifier.
(Amplifier) 22 is given an output current command, thereby outputting a drive current to the motor 23 so that the input speed deviation becomes zero. As a result, the motor 23 outputs the speed command value Nref
Is automatically controlled to a rotation speed that follows the given speed.

As the speed detecting sensor 24, a sensor for providing an analog output such as a tachometer (speed generator) has been mainly used in the past. In many cases, a two-phase pulse output type rotary encoder having high accuracy and also having a function as a motor rotational position detector is used.

A rotary encoder of this type is designed to output a two-phase pulse signal consisting of an A-phase and a B-phase having a phase difference of exactly 1/4 period at every fixed rotation angle of the rotating shaft. The rotation speed can be obtained by measuring the pulse rate of the output pulse, and the rotation direction can be determined from the phase difference between the two-phase pulses.

Next, speed detection in the case where such a rotary encoder is used as the sensor 24 will be described. Normally, in digital control, since speed control is performed at a speed control cycle every fixed sampling period Ts, it is necessary to detect the rotational speed every sampling period Ts.

More specifically, the encoder output pulse count value for each sampling period Ts is taken out, and if the pulse count value changes n times in one sampling period from the previous sampling time to the current sampling time, then Is n / Ts. Therefore, the rotation speed of the encoder, that is, the rotation speed of the motor 23 is calculated from the value n / Ts.

However, in the case of this method, the speed detection value for each sampling period Ts is 1 / Ts in terms of a pulse rate value.
And the resolution greatly decreases as the sampling period Ts is shortened. That is, if the number of pulses per unit rotation angle of the rotary encoder is the same, the number of pulse samplings decreases as the sampling period Ts is shortened.

In recent years, in particular, a motor control system using a vector control method has been widely used. In this case, a high-speed sampling rate with a sampling cycle of milliseconds or less is required in terms of control accuracy and responsiveness. In many cases, a decrease in resolution is a major problem.

Therefore, in order to suppress this decrease in resolution,
A so-called pulse rate type rotation speed detecting device that calculates the rotation speed based on the ratio between the change in the number of encoder output pulses and the pulse input time difference has been conventionally used. Hereinafter, detection of the rotation speed by this method will be described. I do.

First, FIG. 3 is a diagram showing time on the horizontal axis and the count value of the encoder output pulse counter on the vertical axis. Here, time points S0, S1, S2, S3,. At the time of speed sampling, the encoder pulse count value at each sampling time is represented by P0, P1,
P2, P3,..., And the input time of the encoder pulse input immediately before each sampling time is t
0, t1, t2, t3,...

In this pulse rate method, for example, the rotation speed detection value at the sampling point S1 is the ratio of the pulse number change from the timing point S0 to the timing point S1 to the pulse input time change, that is,
(P1−P0) / (t1−t0) is set as the pulse rate, and this is set as the rotation speed.

FIG. 4 shows an example of a speed detector according to this method, which corresponds to, for example, the speed detector 25 in FIG.
The pulse rate (P1-P0) / (t1-t0) is output as a speed detection value Ndet.

From the rotary encoder to the speed detector 25
A and B pulses input to the pulse processing circuit 1
, Where it is first converted into an up-pulse signal U and a down-pulse signal D to which information on the rotation direction is added. The up pulse signal U and the down pulse signal D are supplied to the pulse counter 4, where the number of encoder output pulses is counted up by the up pulse signal U, down counted by the down pulse signal D, and Is output.

At the same time, the pulse processing circuit 1 detects, from each of the input A-phase and B-phase pulses, a pulse edge signal serving as information indicating the input time point of one of the pulses. , To the latch circuit 3 as a latch signal.

The latch circuit 3 functions to latch the count value of the time counter 2 which is counted up by the clock φ of a constant cycle. As a result of operating the pulse edge signal as a latch signal, this latch circuit 3 From time 3, every time each pulse of the encoder output pulse is input, a time T representing the interval is output.

Next, the time T and the pulse count value P are input to the difference circuits 5 and 6, respectively, where the sampling pulse S supplied at every sampling period Ts is supplied.
, The difference from the previous value is obtained, and the difference values ΔT, Δ
P is output. For example, at the sampling time point S1, the difference value ΔT = (t1−t0) and the difference value ΔP = (P1−P0)
At the sampling time S2, the difference value ΔT = (t
2-t1), the difference value ΔP = (P2−P1).

Since the difference values ΔT and ΔP are input to the division circuit 7, a pulse rate ΔP / ΔT, which is a division value of the difference value, is obtained from the output, and is output as a speed detection value Ndet. Therefore, by applying this to the speed control of the motor 23 in FIG. 2, high-accuracy speed control based on a high-resolution speed detection value can be easily obtained.

[0021]

In the above prior art, no consideration is given to noise mixed in the encoder output pulse, and there is a problem that a large error occurs in the speed detection value due to the noise mixed.

In the pulse rate method, as shown in FIG. 3, when there is no noise in the encoder output pulse, a speed detection value with high resolution can be obtained. However, due to the noise, for example, as shown in FIG. For example, when noise is mixed before and after any of the sampling points such as the sampling point S1, a large error occurs in the speed detection value as described below. Here, the change in the number of encoder output pulses from the time point t1 to the time point t2 is due to the inclusion of noise.

As described above, in the pulse rate method, for example, the pulse rate at the sampling point S2 is (P2
−P1) / (t2−t1). At this time, as shown in FIG. 5, a change in the encoder output pulse due to noise appears in a very short time from time t1 to time t2.

Therefore, at this time, the time difference value ΔT
Since (= t2−t1) becomes an extremely small value determined by the width of the noise, the pulse rate represented by ΔP / ΔT is so large that it can be said that it is almost orders of magnitude different from the actual pulse rate. Value. As a result,
A large error occurs in the speed detection value.

In the prior art, even when a speed detection value having a large error occurs, the speed detection value is output as it is, so that the speed detection value has a large error.

If speed control is performed using a speed detection value having such a large error, adverse effects such as a sudden decrease in output torque cannot be avoided. Therefore, conventionally, when such a pulse rate method is adopted, measures such as inserting a delay filter into the detected speed value have been taken.

However, the application of such a delay filter involves a reduction in the response of the speed control. As a result, according to the prior art, the resolution of the speed detection and the response are in a trade-off relationship. In other words, it could not be said that it was a fundamental measure for maintaining the speed detection value with high accuracy.

The present invention has been made in view of the above-mentioned problems, and has as its object to easily obtain a high-accuracy and highly-responsive speed detection value, and to stably apply the present invention to a motor speed control device. It is an object of the present invention to provide a rotational speed detection capable of obtaining a desired control.

[0029]

SUMMARY OF THE INVENTION An object of the present invention is to provide a method of calculating a rotation speed for each predetermined sampling period by using a rotary encoder and calculating a rotation speed at a predetermined sampling cycle based on a ratio between a change in the number of output pulses of the rotary encoder and a pulse input time difference. In the detecting apparatus, limit value calculating means for adding 1 to the pulse difference value detected in each sampling period and outputting the value as a limit value in each sampling period as a limit value is provided. Is achieved such that the rotation speed calculated in the above is limited to the limit value.

Another object of the present invention is to provide a rotation speed detecting apparatus which uses a rotary encoder and calculates a rotation speed at a predetermined sampling period based on a ratio between a change in the number of output pulses of the rotary encoder and a pulse input time difference. A limit value is obtained by adding 1 to the pulse difference value detected for each sampling period, and dividing the value by the period from the time of the output pulse of the low-sufficient encoder that occurred immediately before the previous sampling time to the current sampling time. The present invention is also achieved by providing a limit value calculating means for outputting each sampling period, so that the rotation speed calculated for each sampling period is limited to the limit value.

In the speed detection for each speed control cycle, if it is assumed that n encoder pulses have entered between the previous sampling time and the sampling time, the rotation speed of the encoder in one sampling cycle is n + 1 in one sampling cycle. It is always lower than the rotation speed when a pulse is applied.

The present invention makes use of this relationship.
The limiter that sets the rotational speed of the encoder when the sampling period includes n + 1 pulses as the upper limit of the detected speed value limits the rotational speed detection value of the encoder, and therefore the above object can be achieved. .

[0033]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a rotation speed detecting device according to the present invention will be described in detail with reference to the illustrated embodiments.
FIG. 1 shows a rotation speed detecting device 10 according to an embodiment of the present invention.
In the figure, 8 is a limit value calculator,
Reference numeral 9 denotes a limiter. The other parts of the pulse processing circuit 1, the time counter 2, the latch circuit 3, the pulse counter 4, the difference circuits 5, 6, and the division circuit 7 are the same as those in the prior art described with reference to FIG.

Therefore, the operation from the input of the A-phase pulse and the B-phase pulse supplied from the rotary encoder to the detection of the pulse rate ΔP / ΔT from this is the same as that of the prior art shown in FIG.

In the embodiment shown in FIG. 1, the pulse rate .DELTA.P / .DELTA.T is not output as it is as the speed detection value Ndet, but is temporarily input to the limiter 9 as the speed detection value Nd and the predetermined limit value This is different from the prior art of FIG. 2 in that it is configured to output the speed detection value Ndet after processing by Nlim.

The limit value Nlim by the limiter 9 at this time is configured to be calculated by the limit value calculator 8 from the difference value ΔP between the sampling period Ts and the count value. As shown in FIG. 1, the calculation of the limit value Nlim is characterized by the following equation.

Nlim = (| ΔP | +1) / Ts Here, | ΔP | of the numerator is the absolute value of the difference ΔP of the number of encoder pulses, and Ts of the denominator is the sampling for speed detection as described above. It is a cycle.

As described above, the limiter 9 functions to limit the speed detection value Nd by the limit value Nlim. Specifically, the input speed detection value Nd sets the limit value Nlim to the limit value Nlim. When it exceeds, the limit value Nlim is output as the speed detection value Ndet.

Therefore, the speed detection value Ndet output from the limiter 9 is determined as follows based on the relationship between the speed detection value Nd and the upper limit value Nlim. If Nd <-Nlim → Ndet = -Nlim-Nlim ≤ Nd ≤ Nlim → Ndet = Nd If Nlim <Nd → Ndet = Nlim

Therefore, at any one of the sampling points, for example, as shown in FIG. 5, the speed detection value Nd which is abnormally large at the sampling point S2 is mixed at the sampling point S2 due to the mixing of noise having a narrow width before and after the sampling point S1. Suppose that appears.

On the other hand, at this time, the limit value Nlim by the limit value calculator 8 is 2 / Ts, which is input to the limiter 9. Because the difference value ΔP of the count value at this time, that is, (P2−P
This is because 1) is 1 as is clear from FIG.

As a result, the speed detection value Nd at this time becomes
Depending on the state of the noise, no matter how large the value is, the limiter 9 is actuated when the above state or the state is reached, so that the speed detection value Ndet
Is set to 2 / Ts, which is the limit value Nlim at this time, and the speed detection value Ndet does not become an abnormally large value.

Therefore, according to the rotational speed detecting device 10 shown in FIG. 1, there is no possibility that the detected speed value becomes abnormally large even when noise is mixed, and the detected speed value is output near a normal value. As a result, a speed detection value having high resolution and high responsiveness by the pulse rate method can be easily obtained.

As a result, according to the embodiment of FIG. 1, high-precision control can be easily obtained by applying the present invention to a vector control type motor control device or the like.

Next, as described above, the limit value Nlim in the embodiment of the present invention is given by the following formula: Nlim = (| ΔP | +1) / Ts The reason for the above will be described.

In this case, just because the speed detection value Nd successively given at each sampling time greatly changes during a certain sampling period, it cannot be said that it is not necessarily an abnormal value due to noise. This may be the result of an increase in the rotational speed of the rotary encoder.

Therefore, as the limit value Nlim,
At this time, it is necessary that the given speed detection value Nd can be definitely identified as an abnormal value due to noise or the like and should be as close as possible to the actual speed detection value Ndet.

Accordingly, in the present invention, when the number of encoder pulses between the previous sampling time and the current sampling time is n, one pulse is added to the n pulses, and (n + 1) pulses are added. Then, if the difference value ΔT serving as the denominator is set to the sampling period Ts, the rotation speed represented by n pulses is faster than the rotation speed represented by (n + 1) pulses unless there is an abnormality due to noise or the like. It is noted that the limit value Nlim, which is the upper limit value of the speed detection value, is set based on the above equation, that is, Nlim = (| ΔP | +1) / Ts.

Therefore, according to this embodiment, the upper and lower limits sufficiently close to the actual rotational speed can be set one after another as the limit value Nlim, and as a result, the normal state and the abnormal state can be distinguished extremely. Accurately obtained, it is possible to reliably suppress the occurrence of an abnormal rotation speed detection value due to noise, and to output a speed detection value which is always close to a normal state with little disturbance due to the abnormality.

In the above embodiment, the denominator in the equation for calculating the limit value Nlim is the sampling period Ts.
It has become. However, as is apparent from FIG. 5, a change in the pulse number P due to noise appears at the time point t1.

Therefore, as another embodiment, the limit value Nlim may be calculated by the following equation: Nlim = (| ΔP | +1) / (Ts + Δt). Here, Δt is the time of the previous sampling, that is, S1 in the case of FIG.
The time of the difference between the input pulse points t1 of the immediately preceding encoder pulse, that is, Δt = (S1−t1).

Therefore, according to this other embodiment, the upper and lower limit values closer to the actual rotation speed are set to the limit value Nlim.
As a result, the normal state and the abnormal state can be more accurately distinguished, and a speed detection value with less disturbance due to the abnormality can be easily obtained.

[0053]

According to the present invention, the occurrence of an abnormality detection value in a pulse rate type rotation speed detecting device using a rotary encoder can be reliably suppressed, so that the present invention is applied to a motor control device to achieve high control responsiveness. Originally, highly accurate rotation speed control can be easily obtained.

At this time, even in the case of an abnormality, a speed detection value with little disturbance from the normal state can be directly generated.
A stable rotational speed control state can always be maintained, and a highly reliable motor control device can be easily realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a rotation speed detecting device according to the present invention.

FIG. 2 is a block diagram illustrating an example of a motor control device using a rotation speed detection device.

FIG. 3 is a characteristic diagram for explaining a principle of detecting a rotation speed by a pulse rate method using a rotary encoder.

FIG. 4 is a block diagram illustrating an example of a rotation speed detection device according to the related art.

FIG. 5 is a characteristic diagram for explaining a problem in detecting a rotation speed by a pulse rate method using a rotary encoder.

[Explanation of symbols]

 Reference Signs List 1 pulse processing circuit 2 counter (for time measurement) 3 latch circuit (for time data latch) 4 pulse counter 5 difference circuit (for time difference) 6 difference circuit (for pulse difference) 7 division circuit 8 limit value calculator 9 limiter 10 Rotation speed detection device according to one embodiment of the invention 20 Speed control device 21 Speed controller 22 Current amplifier 23 Motor 24 Speed sensor 25 Rotation speed detector 26 Speed difference subtractor

Claims (2)

[Claims] 1. A rotation speed detecting device which uses a rotary encoder and calculates a rotation speed for each predetermined sampling period based on a ratio between a change in the number of output pulses of the rotary encoder and a pulse input time difference. The pulse difference value detected at
Is provided, and a limit value calculating means for outputting a value obtained by dividing by the sampling period as a limit value for each sampling period is provided, and the rotation speed calculated for each sampling period is limited to the limit value. A rotation speed detecting device, characterized in that: 2. A rotation speed detecting device which uses a rotary encoder and calculates a rotation speed for each predetermined sampling cycle based on a ratio between a change in the number of output pulses of the rotary encoder and a pulse input time difference. The pulse difference value detected at
Limit value calculation that outputs a value obtained by dividing the period from the time point of the output pulse of the low-sufficient encoder that occurred immediately before the previous sampling time point to the current sampling time point as a limit value for each sampling period. Means, wherein a rotation speed calculated for each sampling period is limited to the limit value.