JP4930919B2 - Rotational position signal processor - Google Patents

Description

  The present invention relates to a rotational position signal processing apparatus that processes a signal from a rotational detection sensor such as a resolver and a synchro, for example, and in particular, a correction means corrects an output level of a digital position signal, and updates a correction position signal. When the interpolation means predicts and the corrected position signal is not updated at the prediction timing, the interpolation means outputs the predicted position signal, so that the output distribution of the position signal can be made denser, and the control accuracy of the control target The present invention relates to a new improvement for improving the above.

  As this type of rotational position signal processing device that has been used in the past, for example, the rotational position signal processing device disclosed in Patent Document 1 is used. FIG. 5 is a block diagram showing a conventional rotational position signal processing device 2. In the figure, a rotation detection sensor 1 including, for example, a resolver and a synchro is attached to a monitoring target (not shown) such as a motor. A drive control means 3 is connected to the rotation detection sensor 1 via a rotation position signal processing device 2, and the rotation position signal processing device 2 is, for example, an R / D converter or an S / D converter. It consists of a signal converter 20 and an amplifier 21. The signal converter 20 converts the analog position signal 1 a from the rotation detection sensor 1 into a digital position signal 20 a and inputs the digital position signal 20 a to the amplifier 21 and the drive control means 3. The amplifier 21 amplifies the digital position signal 20a and feeds it back to the rotation detection sensor 1 as an excitation signal 21a. The drive control means 3 performs drive control of the monitoring target based on the digital position signal 20a.

Japanese Patent Laid-Open No. 11-118520

  By the way, the digital position signal 20a output from the rotational position signal processing device 2 includes a position error inherent to the sensor due to manufacturing and a position error due to the cable length at the time of sensor installation. Even if the detection sensor 1 rotates at a constant speed, the output level of the digital position signal 20a and the rotation angle of the rotation detection sensor 1 are not in direct proportion, and the amount of change and the change timing of the output level are coarse (variable). Arise. Therefore, in the conventional rotational position signal processing device 2 as described above, it is difficult to accurately monitor the rotation angles at equal intervals, and there is a possibility that an error occurs in the drive control of the monitoring target.

  The present invention has been made in order to solve the above-described problems, and an object of the present invention is to make it possible to make the output distribution of the position signal denser and improve the control accuracy of the controlled object. It is about improvement.

  The rotational position signal processing apparatus according to the present invention includes a signal converter connected to a rotation detection sensor, a rotation time measuring means connected to the signal converter, the signal converter and the rotation time measuring means. Connected correction means, correction amount calculation section and correction processing section that constitute the correction means, required rotation time measurement means and interpolation means connected to the correction means, the correction means and the interpolation means And a resolution storage means connected to the sensor, wherein the signal converter converts an analog position signal from the rotation detection sensor into a digital position signal, and the required rotation time measuring means is based on the digital position signal. The time required for one rotation of the rotation detection sensor is measured, and the resolution storage means stores the output level resolution and change timing resolution of the digital position signal, and calculates the correction amount. Part) stores the digital position signal during one rotation of the rotation detection sensor when the rotation detection sensor detects that the rotation detection sensor is rotating at a constant speed based on the time required for one rotation. Obtaining and storing a correction amount for removing an error of the digital position signal based on the time required for one rotation, the output level and change timing of the stored digital position signal, and the output level resolution; When the correction processing unit detects that the rotation detection sensor is rotating at a constant speed based on the time required for one rotation, the correction processing unit uses the correction amount of the correction amount calculation unit to output from the signal converter. A correction position signal is generated by correcting the digital position signal, and the correction position signal and correction execution information are input to the interpolation means, Predicts the timing at which the correction position signal from the correction processing unit is updated based on the time required for one rotation and the change timing resolution when the correction execution information is detected, and the correction position signal Is estimated based on the time required for one rotation, the prediction timing, and the output level resolution, and the predicted change when the correction position signal is not updated at the prediction timing. In this configuration, a predicted position signal is output using a level.

  Further, the interpolation means predicts the prediction timing by T ′ / NT, where T ′ is the time required for one rotation when the correction execution information is detected, and NT is the change timing resolution, and the prediction timing is d, where the output level resolution is NL, the predicted change level is predicted by (d / T ′) × NL, the predicted change level is EA, the output level of the corrected position signal before update is RO, and RO + EA The output level of the predicted position signal is determined.

  The correction amount calculation unit sets the time required for one rotation when the rotation detection sensor is rotating at a constant speed as T, the output level of the stored digital position signal as L, and the output level. The correction amount is obtained by {(t / T) × NL} −L where t is the corresponding change timing and NL is the output level resolution.

  According to the rotational position signal processing apparatus of the present invention, the correction unit corrects the output level of the digital position signal, and the interpolation unit predicts the update timing of the correction position signal. Since the interpolation means outputs the predicted position signal when it is not updated, the output distribution of the position signal can be made denser and the control accuracy of the controlled object can be improved.

  Further, the interpolation means predicts the prediction timing by T ′ / NT, where T ′ is the time required for one rotation when the correction execution information is detected, and NT is the change timing resolution, and the prediction timing is d, where the output level resolution is NL, the predicted change level is predicted by (d / T ′) × NL, the predicted change level is EA, the output level of the corrected position signal before update is RO, and RO + EA Since the output level of the predicted position signal is determined, the predicted position signal can be output more reliably, and the control accuracy of the controlled object can be improved more reliably.

  In addition, the correction amount calculation unit T indicates the time required for one rotation when it is determined that the rotation detection sensor is rotating at a constant speed, and Ln indicates the output level of the stored digital position signal. Since the change amount corresponding to the level L is t and the output level resolution is NL, the correction amount is obtained by {(t / T) × NL} -L, so that the output level of the digital position signal can be corrected more reliably. Thus, the control accuracy of the controlled object can be improved more reliably.

The best mode for carrying out the present invention will be described below with reference to the drawings.
Embodiment 1 FIG.
FIG. 1 is a block diagram showing a rotational position signal processing device 2 according to Embodiment 1 of the present invention. FIG. 2 is a graph showing the digital position signal 20a output from the signal converter 20 of FIG. FIG. 3 is a graph showing the position signal after the correction process and the interpolation process are performed by the correction unit 24 and the interpolation unit 25 of FIG. The same or equivalent parts as those of the conventional rotational position signal processing apparatus will be described using the same reference numerals.

  In FIG. 1, a rotation detection sensor 1 including, for example, a resolver and a synchro is attached to a monitoring target (not shown) such as a motor, and outputs an analog position signal 1a according to a rotation angle. The rotation detection sensor 1 is connected to a drive control means 3 via a rotation position signal processing device 2, and the rotation position signal processing device 2 processes the analog position signal 1a, and the drive control means 3 performs the rotation control. Based on the signals 20a, 23a to 25a from the position signal processing device 2, drive control of the monitoring target is performed.

  The rotational position signal processing device 2 comprises a signal converter 20, an amplifier 21, a required rotation time measuring means 22, a speed detecting means 23, a correcting means 24, an interpolating means 25, and a resolution storing means 26.

  The signal converter 20 includes, for example, an R / D converter and an S / D converter, and is connected to the rotation detection sensor 1 to convert the analog position signal 1a into a digital position signal 20a. The amplifier 21 amplifies the digital position signal 20a and feeds back the amplified signal to the rotation detection sensor 1 as an excitation signal 21a.

  The digital position signal 20a is a signal whose output level changes stepwise according to the rotation angle. A plurality of points (P0, P1, P2... Pn) plotted in FIG. 2 indicate the output level (L0, L1... Ln) and the change timing (t0, t1... Tn) of the digital position signal 20a. ). The change timing is the elapsed time from when the rotation angle is 0 ° (reference timing). This digital position signal 20a includes a position error inherent to the sensor due to manufacturing, a position error due to the cable length at the time of sensor installation, and the like, even if the rotation detection sensor 1 rotates at a constant speed. As shown in FIG. 2, the output level of the digital position signal 20a and the rotation angle of the rotation detection sensor 1 are not directly proportional. However, the reproducibility of these position errors is extremely high. When the rotation detection sensor 1 is rotating at a constant speed, the digital position signal 20a having the same output level is output at the same change timing every rotation.

  Returning to FIG. 1, the rotation time measuring means 22 is connected to the signal converter 20, and the time required for the rotation detection sensor 1 to rotate 360 ° (referred to as one rotation required time) is Measurement is performed based on the digital position signal 20a. The required rotation time measuring means 22 will be described later with reference to the drawings.

  The speed detection means 23 is connected to the required rotation time measuring means 22, and obtains the rotation speed of the rotation detection sensor 1 based on the required one rotation time 22 a measured by the required rotation time measuring means 22. Further, the speed detection means 23 inputs the obtained rotation speed to the drive control means 3 as a speed signal 23a.

  The resolution storage means 26 is connected to the correction means 24 and the interpolation means 25, and stores an output level resolution 26a and a change timing resolution 26b of the digital position signal 20a. The output level resolution 26a and the change timing resolution 26b are used for the operation of the correction means 24 and the interpolation means 25. The output level resolution 26a is an ideal value when the rotation detection sensor 1 is rotated by a unit angle. The output level change amount is shown, and the change timing resolution 26b shows the time required for updating the output level. The output level resolution 26a and the change timing resolution 26b are set values input from the outside.

  The correction unit 24 is connected to the signal converter 20, the required rotation time measurement unit 22, and the resolution storage unit 26. The correction unit 24 includes a correction amount calculation unit 240 and a correction processing unit 241.

  The correction amount calculation unit 240 determines whether or not the rotation detection sensor 1 is rotating at a constant speed based on the time required for one rotation 22a. The digital position signal 20a is stored during one rotation of the detection sensor 1. That is, when it is determined that the correction amount calculation unit 240 rotates at a constant speed, the correction amount calculation unit 240 monitors the output level of the digital position signal 20a shown in FIG. 2 and outputs the current level every time the output level is updated. The level and change timing are stored in order. The correction amount calculation unit 240 creates the position data table shown in Table 1 by storing the output level and the change timing in order.

  In addition, the correction amount calculation unit 240 calculates the correction amount for removing the error included in the digital position signal 20a, the time required for one rotation 22a, the output level of the stored digital position signal 20a, and the change timing. It is determined based on (position data table of Table 1) and the output level resolution 26a. Specifically, the time required for one rotation 22a when it is determined that the rotation detection sensor 1 is rotating at a constant speed is T, the output level of the stored digital position signal 20a is L, and the output level L is The correction amount calculation unit 240 obtains the correction amount by {(t / T) × NL} −L, where t is the corresponding change timing and the output level resolution 26a is NL. That is, by calculating t / T, for example, an angle index indicating at which angle the digital position signal 20a at the point P1 in FIG. 2 is output at 360 ° is obtained and output to the angle index. The ideal output level at the angle is obtained by multiplying the level resolution 26a, and the correction amount is obtained by subtracting the actual output level from the ideal output level.

  The correction amount calculation unit 240 sequentially calculates the correction amounts of the digital position signal 20a for one rotation, and creates a correction amount storage table using the angle index as key data as shown in Table 2.

  The correction amount calculation operation of the correction amount calculation unit 240 is an operation performed independently of the correction processing operation by the correction processing unit 241 described later, and is repeated when the rotation detection sensor 1 is rotating at a constant speed. Done. That is, the correction amount calculation unit 240 repeatedly performs the correction amount calculation operation to accumulate a plurality of correction amounts in the correction amount storage table, and when there are a plurality of correction amounts of the same angle index, the average value thereof is calculated. Ask.

  The correction processing unit 241 determines whether or not the rotation detection sensor 1 is rotating at a constant speed based on the time required for one rotation 22a, and the correction amount storage table (see Table 2) is created. It is determined whether or not. When it is determined that the rotation detection sensor 1 is rotating at a constant speed and the correction amount storage table is created, the correction processing unit 241 determines that the correction processing of the digital position signal 20a is possible. Then, the correction position signal 24a is generated by correcting the digital position signal 20a from the signal converter 20 using the correction amount. Specifically, the correction processing unit 241 obtains an angle index of the digital position signal 20a from the signal converter 20, and adds a correction amount corresponding to the calculated angle index to the output level of the digital position signal 20a. Thus, the corrected position signal 24a is generated. At this time, if the correction amount corresponding to the calculated angle index is not registered in the correction amount storage table, the correction processing unit 241 uses the correction amount of the closest angle index among the registered angle indices. . The correction position signal 24a is indicated by a black dot in FIG. 3, and by using the correction amount (r1, r2,... Rn), the output of the digital position signal 20a is directly proportional to the change in the rotation angle. The level is corrected. The correction processing unit 241 inputs the generated correction position signal 24 a and correction execution information 24 b to the interpolation unit 25.

  On the other hand, when the correction processing unit 241 determines that the correction processing of the digital position signal 20a is impossible, the correction processing unit 241 inputs the digital position signal 20a to the drive control unit 3 without performing the correction processing.

  The interpolation means 25 is connected to the required rotation time measuring means 22 and the correction means 24. When the correction execution information 24b from the correction means 24 is detected, an interpolation operation of the correction position signal 24a is performed. I do.

  The interpolation means 25 predicts the timing at which the correction position signal 24a from the correction means 24 is updated based on the one-revolution required time 22a and the change timing resolution 26b. Specifically, the prediction timing is predicted by T ′ / NT, where T ′ is the time required for one rotation 22 t when the correction execution information 24 b is detected, and NT is the change timing resolution 26 b.

  The interpolation means 25 predicts a change level when the correction position signal 24a is updated based on the one-revolution required time 22a, the prediction timing, and the output level resolution 26a. Specifically, the change level is predicted by (d / T ′) × NL, where d is the prediction timing and NL is the resolution of the output level.

  Furthermore, the interpolation means 25 outputs the predicted position signal 25a using the predicted change level when the corrected position signal 24a is not updated at the prediction timing. Specifically, assuming that the predicted change level is EA and the output level of the corrected position signal 24a before update is RO, the output level is determined by RO + EA, and the predicted position signal 25a of the output level is output. The predicted position signal 25a is indicated by white points (c4, c5,..., Cn) in FIG. 3, and compensates for the update when the correction position signal 24a is delayed.

  Next, FIG. 4 is a block diagram showing in more detail the configuration of the required rotation time measuring means 22 of FIG. As described above, the digital position signal 20a includes a position error inherent to the sensor due to manufacturing, a position error due to the cable length at the time of sensor installation, and the like. The reproducibility of these position errors. Is extremely expensive. By utilizing this high reproducibility, the time required for one rotation can be accurately measured. However, if the time required for one rotation is measured for each rotation, the measurement frequency becomes low. Therefore, the time required for rotation measuring means 22 of this embodiment is the same as the rotation detection sensor 1 is rotated every predetermined angle. The measurement of the time required for one rotation 22a is started.

  In the figure, the required rotation time measuring means 22 includes a timing generation circuit 220, a timer 221, a plurality of latch circuit sections 222, a plurality of subtraction circuits 223, and a selection circuit 224. The timer 221 includes a crystal resonator 221a that emits clock pulses and an UP counter 221b that counts the clock pulses. The UP counter 221b counts clock pulses, and outputs timer information 221c that indicates the accuracy to the microsecond (μs) unit. Each latch circuit unit 222 stores (latches) the timer information 221c from the UP counter 221b.

  The timing generation circuit 220 detects the rotation angle of the rotation detection sensor 1 based on the digital position signal 20a, and each time the rotation detection sensor 1 rotates by an angle that is equally divided by 360 degrees, each latch circuit section 222 is detected. The start time storage command 220a is sequentially input. Further, the timing generation circuit 220 inputs the end time storage command 220b to the latch circuit portion 222 when the rotation detection sensor 1 makes one rotation after the start time storage command 220a is input to any of the latch circuit portions 222. To do. The latch circuit unit 222 includes one set each of a start time latch circuit 222a and an end time latch circuit 222b. Timer information 221c when the start time storage command 220a is input (one rotation of the rotation detection sensor 1 is started). Timer information 221c) is stored in the start time latch circuit 222a, and timer information 221c when the end time storage command 220b is input (timer information 221c when one rotation of the rotation detection sensor 1 ends) is stored. The end time latch circuit 222b stores it.

  More specifically, the timing generation circuit 220 sequentially inputs a start time storage command 220a to each latch circuit unit 222 every time the digital position signal 20a from the signal converter 20 is displaced by a predetermined constant level. . Here, since the digital position signal 20a includes a peculiar error due to the combination of the rotation detection sensor 1 and the signal converter 20, each time the rotation detection sensor 1 rotates by 360 degrees equally divided, it is precisely performed. The digital position signal 20a is not output. That is, the timing generation circuit 220 sequentially inputs the start time storage command 220a to each latch circuit unit 222 every time the rotation detection sensor 1 rotates by an angle obtained by dividing 360 degrees substantially equally. Further, the timing generation circuit 220 is the same as when the level of the digital position signal 20a is input to the latch circuit unit 222 after the start time storage command 220a is input to any of the latch circuit units 222. When the level is reached, it is detected that the rotation detection sensor 1 has made one rotation, and an end time storage command 220 b is input to the latch circuit unit 222.

  Each subtraction circuit 223 obtains the time required for one rotation of the rotation detection sensor 1 based on the timer information 221c stored in each start time latch circuit 222a and each end time latch circuit 222b. The timing generation circuit 220 inputs the rotation angle information 220c to the selection circuit 224. Based on the rotation angle information 220c from the timing generation circuit 220, the selection circuit 224 selects the latest one rotation required time 22a from the one rotation required time 22a calculated by each subtraction circuit 223, and detects the speed. Input to the means 23, the correction means 24, and the interpolation means 25.

  In such a rotational position signal processing device 2, the correction means 24 corrects the output level of the digital position signal 20a, and the interpolation means 25 predicts the update timing of the correction position signal 24a, and the correction is performed at the prediction timing. Since the interpolation means 25 outputs the predicted position signal 25a when the position signal 24a is not updated, the output distribution of the position signal can be made denser and the control accuracy of the controlled object can be improved.

  The interpolation means 25 predicts the prediction timing by T ′ / NT, where T ′ is the time required for one rotation when the correction execution information 24b is detected, T ′, and the change timing resolution 26b is NT. The prediction timing is d, the output level resolution 26a is NL, the prediction change level is predicted by d / T ′ × NL, the prediction change level is EA, and the output level of the correction position signal before update is RO. Since the output level of the predicted position signal 25a is determined by RO + EA, the predicted position signal can be output more reliably and the control accuracy of the controlled object can be improved more reliably.

  Further, the correction amount calculation unit 240 sets the time required for one rotation 22a when the rotation detection sensor 1 is rotating at a constant speed to T, sets the output level of the stored digital position signal 20a to L, Since the change amount corresponding to the output level is t and the output level resolution 26a is NL, the correction amount is obtained by {t / T × NL} -L, so that the output level of the digital position signal can be corrected more reliably. Thus, the control accuracy of the controlled object can be improved more reliably.

It is a block diagram which shows the rotational position signal processing apparatus by Embodiment 1 of this invention. It is a graph which shows the digital position signal output from the signal converter of FIG. It is a graph which shows the position signal after a correction process and an interpolation process were performed by the correction means and the interpolation means of FIG. It is a block diagram which shows the structure of the rotation required time measurement means of FIG. 1 in detail. It is a block diagram which shows the conventional rotational position signal processing apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Rotation detection sensor 1a Analog position signal 2 Rotation position signal processing apparatus 3 Drive control means 20 Signal converter 20a Digital position signal 22 Rotation required time measurement means 22a 1 Rotation required time 24 Correction means 24a Correction position signal 24b Correction execution information 240 Correction Quantity calculation unit 241 Correction processing unit 25 Interpolation unit 25a Predicted position signal 26 Resolution storage unit 26a Output level resolution 26b Change timing resolution

Claims (3)

A signal converter (20) connected to the rotation detection sensor (1);
A rotation time measuring means (22) connected to the signal converter (20);
Correction means (24) connected to the signal converter (20) and the required rotation time measuring means (22);
A correction amount calculation unit (240) and a correction processing unit (241) constituting the correction means (24);
An interpolation means (25) connected to the rotation required time measurement means (22) and the correction means (24);
A resolution storage means (26) connected to the correction means (24) and the interpolation means (25);
The signal converter (20) converts the analog position signal (1a) from the rotation detection sensor (1) into a digital position signal (20a),
The required rotation time measuring means (22) measures a required rotation time (22a) of the rotation detection sensor (1) based on the digital position signal (20a),
The resolution storage means (26) stores an output level resolution (26a) and a change timing resolution (26b) of the digital position signal (20a),
When the rotation amount detection unit (240) detects that the rotation detection sensor (1) is rotating at a constant speed based on the time required for one rotation (22a), the rotation detection sensor (1) The digital position signal (20a) during one rotation of the digital signal is stored, the time required for one rotation (22a), and the output level (L0, L1, L2,...) Of the stored digital position signal (20a). ... Ln) and a correction amount (r1) for removing an error of the digital position signal (20a) based on the change timing (t0, t1, t2,... Tn) and the output level resolution (26a). , R2,... Rn) and store them,
When the correction processing unit (241) detects that the rotation detection sensor (1) is rotating at a constant speed based on the time required for one rotation (22a), the correction amount calculation unit (240) A correction position signal (24a) is generated by correcting the digital position signal (20a) from the signal converter (20) using the correction amount of the correction position signal, and the correction position signal (24a) and correction execution information. (24b) is input to the interpolation means (25),
When the interpolation means (25) detects the correction execution information (24b), the interpolation means (25) determines the timing at which the correction position signal (24a) from the correction processing unit (241) is updated as the time required for one rotation ( 22a) and the change timing resolution (26b), and the change level when the correction position signal (24a) is updated, the time required for one rotation (22a), the prediction timing, Predicting based on the output level resolution (26a), and outputting the predicted position signal (25a) using the predicted change level when the corrected position signal (24a) is not updated at the prediction timing. Rotational position signal processing device. The interpolation means (25)
When the correction execution information (24b) is detected, the one-revolution required time (22a) is T ′, the change timing resolution (26b) is NT, and the prediction timing is predicted by T ′ / NT,
The prediction change level is predicted by (d / T ′) × NL, where d is the prediction timing and the output level resolution (26a) is NL.
The rotational position signal processing according to claim 1, wherein the output level of the predicted position signal (25a) is determined by RO + EA, where the predicted change level is EA and the output level of the corrected position signal before update is RO. apparatus. The correction amount calculation unit (240)
The time required for one rotation (22a) when it is determined that the rotation detection sensor (1) is rotating at a constant speed is T, the output level of the stored digital position signal (20a) is L, and the output level is The correction amount is obtained by {(t / T) × NL} -L, where t is the corresponding change timing and NL is the output level resolution (26a). Rotational position signal processing device.

Images