JP5943670B2 - Encoder device and position data abnormality detection method - Google Patents

Description

  The present invention relates to an encoder device for detecting a rotation angle or the like of a rotating body. The present invention also relates to an abnormality detection method for position data in an encoder device for detecting a rotation angle or the like of a rotating body.

  Conventionally, an encoder for detecting the rotation angle of a motor and an AC servo driver for controlling the motor based on position data detected by the encoder are widely used. The position data detected by the encoder may become erroneous position data due to a bit error due to electromagnetic noise or the like when it is sent from the encoder to the AC servo driver. If the position data sent to the AC servo driver is incorrect position data, the behavior of the rotating motor becomes unstable, and there is a possibility that the rotation unevenness and vibration of the motor may occur.

  Therefore, conventionally, it is detected that the position data detected by the encoder has become incorrect position data due to the influence of electromagnetic noise or the like sent from the encoder to the AC servo driver (that is, the position data is detected to be abnormal). Thus, there has been proposed a bit error detection / estimation method for detected position data that makes it possible to estimate correct position data (see, for example, Patent Document 1).

  In the bit error detection / estimation method described in Patent Document 1, the maximum position change amount during the sampling period h is calculated from the maximum rotational speed of the motor and the encoder resolution N at the time of use, and the position data at the time of this sampling is calculated. pos # s1, which is the absolute value of the change amount (difference) between pos # b and the position data pos # a at the previous sampling, and pos #, which is the absolute value of the change amount between the encoder resolution N and pos # s1. s2 is calculated. When pos # s1 and pos # s2 exceed the maximum position change amount during the sampling period h, it is determined that there is a large bit error in pos # b. That is, by determining whether pos # s1 and pos # s2 exceed the maximum position change amount during the sampling period h, the position data is influenced by the influence of electromagnetic noise or the like when sent from the encoder to the AC servo driver. It is detected that an abnormality has occurred.

Further, in this bit error detection / estimation method, when it is detected that an abnormality has occurred in the position data, the position data pos # a at the previous sampling, the speed Vb during the previous sampling period h, and the sampling period The correct position data pos # bs is estimated based on the following equation using h.
pos # bs = pos # a + Vb × h
The AC servo driver controls the motor based on the estimated position data pos # bs.

JP 2004-37246 A

  However, in the bit error detection / estimation method described in Patent Document 1, it is detected that an abnormality has occurred in the position data based only on pos # s1 and pos # s2, which are changes in position data. It cannot be said that the position data abnormality detection accuracy is high. If the position data abnormality detection accuracy is low, the position data cannot be detected even though there is an abnormality in the position data, so the correct position data cannot be estimated and it is difficult to rotate the motor smoothly. There is a risk of becoming.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an encoder device that can improve the accuracy of abnormality detection of position data before processing (position data before processing) output from an encoder. Another object of the present invention is to provide a position data abnormality detection method capable of improving the accuracy of abnormality detection of pre-processing position data.

In order to solve the above problems, an encoder apparatus according to the present invention includes an encoder, and a position data processing unit that processes post-processing position data output from the encoder and generates post-processing position data. The first floor difference value data before processing, which is the difference value of the position data before processing acquired every predetermined sampling cycle, is calculated, and the second floor difference value, which is the difference value of the first floor difference value data before processing, for each sampling cycle. When the first floor differential value data before processing is within the predetermined first reference range and the second floor differential value data is within the predetermined second reference range, the position data before processing is normal. The position data before processing is abnormal when it is determined that there is, and the first floor differential value data before processing is out of the first reference range and / or when the second floor differential value data is out of the second reference range. In Determination and, further, in the sampling period, the previous, and after the position data generated processed this time, calculates a first-order difference value data before and after the treatment is a difference value between the obtained pretreated position data pretreatment If it is determined that the position data is normal, the current post-processing position data is generated using the first-order difference data calculated before and after the current processing in the sampling period, and the pre-processing position data is determined to be abnormal. Then, in the sampling period, the average value of the first-order difference data before and after the predetermined number of processes calculated up to the previous time is estimated as the first-order difference value data before and after the current process, and the average value is used after the current process. It is characterized by generating position data .

  In the encoder device of the present invention, the position data processing unit calculates the first-order difference value data and the second-order difference value data before processing, and the first-order difference value data before processing is out of the first reference range; and / Or When the second-order difference value data is out of the second reference range, it is determined that the pre-processing position data is abnormal. That is, in the present invention, the pre-processing position data is used by using the second-order difference value data that is the difference value of the pre-processing first-order difference value data in addition to the pre-processing first-order difference value data that is the difference value of the pre-processing position data. An abnormality is detected. Therefore, in the present invention, it is possible to increase the accuracy of detecting the abnormality of the pre-processing position data as compared with the case of detecting the pre-processing position data abnormality using only the pre-processing first-order difference value data.

Further, in the present invention , the position data processing unit outputs the first and second floor difference value data before and after processing, which is a difference value between the previously generated post-processing position data and the current pre-processing position data acquired this time in the sampling period. When calculating and determining that the pre-processing position data is normal, the current post-processing position data is generated using the first-order differential value data before and after the calculation in the sampling period, and the pre-processing position data Is determined to be abnormal in the sampling period, the average value of the first-order difference data before and after the processing of the predetermined number calculated up to the previous time is estimated as the first-order difference value data before and after the current process, and this average value is This is used to generate the post-processing position data .

Therefore, if the pre-processing position data is normal, the current post-processing position data is generated based on the pre-processing position data acquired this time. For example, when the rotation angle of the motor is detected by the encoder Therefore, it is possible to prevent the rotation characteristics of the motor from being adversely affected. Also, if the processed front position data is abnormal, on the basis of the average value of the first-order difference value data before and after processing of the calculated predetermined number up to the previous time, since the current process after the position data is generated, for example, When the rotation angle of the motor is detected by the encoder, a phenomenon that the behavior of the motor becomes unstable due to post-processing position data generated based on abnormal pre-processing position data (specifically, It is possible to prevent the occurrence of motor vibration or vibration).

  In the present invention, for example, when the position data processing unit determines that the pre-processing position data is normal, the first-order difference value data before and after the process calculated this time and the before and after the process calculated up to the previous time in the sampling period If the sum of the first-order difference value data is calculated, the current post-processing position data is generated, and it is determined that the pre-processing position data is abnormal, the predetermined number of processes calculated up to the previous time in the sampling period The sum of the average value of the previous and subsequent first-order difference value data and the previous and subsequent first-order difference value data calculated up to the previous time is calculated, and the current post-processing position data is generated.

In the present invention, the position data processing unit determines that the pre-processing position data is normal, and if the first-order difference data before and after the processing is within a predetermined third reference value, the current calculation is performed in the sampling period. Even if the post-processing position data is generated using the first and second floor difference value data before and after processing, and the pre-processing position data is determined to be normal, the first and second floor difference value data before and after processing is set to the third reference value. If it exceeds, in the sampling period, the average value of the first-order difference value data before and after the processing of the predetermined number calculated up to the previous time is estimated as the first-order difference value data before and after the current process, and using this average value, It is preferable to generate the post-processing position data.

  If the first floor difference value data before and after the process calculated this time exceeds the third reference value (that is, if the first floor difference value data before and after the process calculated this time is larger than the assumed value), The pre-processing position data acquired this time is assumed to be abnormal, and the post-processing position data generated last time is an estimated value generated using the average value of the first-order difference data before and after the predetermined number of processes. It is assumed that the estimated post-processing position data is a value far from the normal pre-processing position data. Calculated based on the estimated post-processing position data at the previous sampling when the estimated post-processing position data at the previous sampling is significantly different from the normal pre-processing position data. If the current post-processing position data is generated based on the first-stage difference value data before and after the current processing, the generated current post-processing position data may become a value far from the correct post-processing position data. There is.

  Therefore, when the first-stage difference value data before and after the processing calculated this time exceeds a predetermined reference value, the average value of the predetermined number of first-order difference value data before and after the process calculated up to the previous time in the sampling period, Estimating the first-order difference data before and after the current process, and using this average value to generate the current post-process position data, the estimated post-process position data is far from the correct post-process position data. Can be prevented. That is, the difference between the estimated post-processing position data and the correct post-processing position data can be suppressed. Accordingly, the estimated post-processing position data can be made to follow the normal pre-processing position data, and the estimated post-processing position data can be prevented from greatly deviating from the normal pre-processing position data. It becomes possible. As a result, for example, when the rotation angle of the motor is detected by the encoder, the behavior of the motor can be stabilized.

In the present invention, for example, the position data processing unit determines that the pre-processing position data is normal, and the first-order difference value data before and after processing is within the third reference value, this time in the sampling period, To calculate the sum of the first-order difference value data before and after processing and the first-order difference value data before and after processing calculated up to the previous time, and generate the current post- processing position data. Even if it is determined that there is a first-order difference value data before and after processing exceeds the third reference value, an average value of the predetermined number of first-order difference data before and after processing calculated up to the previous time in the sampling period, The sum of the difference value data before and after processing calculated up to the previous time is calculated, and the current post-processing position data is generated.

In order to solve the above problems, the position data abnormality detection method of the present invention includes an encoder and a position data processing unit that processes post-processing position data output by the encoder and generates post-processing position data. A first-order difference value calculating step for calculating pre-processing first-order difference value data, which is a difference value of pre-processing position data acquired every predetermined sampling period, in a position data abnormality detection method in an encoder device; and sampling A second-order difference value calculation step for calculating second-order difference value data that is a difference value of the first-order difference value data before processing for each cycle, and whether or not the first-order difference value data before processing is within a predetermined first reference range a first evaluation step of evaluating whether, second-order difference data and a second evaluation step of evaluating whether within a predetermined second reference range, the first evaluation step odor It is detected that the pre-processing position data is normal when the pre-processing first floor differential value data is within the first reference range and the second evaluation step data is within the second reference range in the second evaluation step. When the first- stage differential value data before processing is out of the first reference range in the first evaluation step , and / or when the second-order difference value data is out of the second reference range in the second evaluation step. The pre-processing position data is detected to be abnormal , and in the sampling period, the first floor before and after the processing is the difference value between the previously generated post-processing position data and the pre-processing position data acquired this time. When the difference value data is calculated and when the pre-processing position data is detected to be normal, the first-order difference value data before and after the processing calculated this time is used in the sampling period. When the post-processing position data of this time is generated and it is detected that the pre-processing position data is abnormal, the average value of the predetermined number of first-order differential value data before and after the processing calculated up to the previous time in the sampling period, It is estimated that the first-stage differential value data before and after the current process, and using the average value, the step of generating the current post-process position data is provided .

In the position data abnormality detection method of the present invention, pre-processing first-order difference value data is calculated in the first-floor difference value calculating step, and second-order difference value data is calculated in the second-floor difference value calculating step. When the first-order difference value data is out of the first reference range and / or when the second-order difference value data is out of the second reference range, it is determined that the pre-processing position data is abnormal. . That is, in the present invention, the pre-processing position data is used by using the second-order difference value data that is the difference value of the pre-processing first-order difference value data in addition to the pre-processing first-order difference value data that is the difference value of the pre-processing position data. An abnormality is detected. Therefore, in the present invention, it is possible to increase the accuracy of detecting the abnormality of the pre-processing position data as compared with the case of detecting the pre-processing position data abnormality using only the pre-processing first-order difference value data. In the position data abnormality detection method of the present invention, the first and second floor difference value data before and after processing, which is a difference value between the previously generated post-processing position data and the current pre-processing position data acquired this time in the sampling period. When the position data before processing is detected to be normal, the current post-processing position data is generated using the first-order difference value data before and after the processing that is calculated this time in the sampling period. When the pre-processing position data is detected to be abnormal, the average value of the predetermined number of first-order difference data before and after processing calculated up to the previous time in the sampling period is used as the first-order difference value data before and after the current processing. And generating the post-processing position data this time using the average value. Therefore, if the pre-processing position data is normal, the current post-processing position data is generated based on the pre-processing position data acquired this time. For example, when the rotation angle of the motor is detected by the encoder Therefore, it is possible to prevent the rotation characteristics of the motor from being adversely affected. In addition, if the pre-processing position data is abnormal, the current post-processing position data is generated based on the average value of the predetermined number of first-order differential data before and after processing calculated up to the previous time. When the rotation angle of the motor is detected by the phenomenon, the behavior of the motor becomes unstable due to the post-processing position data generated based on the abnormal pre-processing position data (specifically, the motor Occurrence of vibration and vibration).

  As described above, in the encoder device of the present invention, it is possible to improve the accuracy of detecting abnormality in the pre-processing position data output from the encoder. Further, according to the position data abnormality detection method of the present invention, it is possible to improve the accuracy of the abnormality detection of the pre-processing position data.

It is a block diagram for demonstrating schematic structure of the encoder apparatus concerning embodiment of this invention. It is a figure for demonstrating schematic structure of the encoder shown in FIG. It is a block diagram for demonstrating schematic structure of the position data processing part shown in FIG. It is a flowchart for demonstrating the abnormality detection method of the position data before a process in the position data process part shown in FIG. It is a flowchart for demonstrating the production | generation method of the post-processing position data in the position data processing part shown in FIG.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Configuration of encoder device)
FIG. 1 is a block diagram for explaining a schematic configuration of an encoder apparatus 1 according to an embodiment of the present invention. FIG. 2 is a diagram for explaining a schematic configuration of the encoder 3 shown in FIG. FIG. 3 is a block diagram for explaining a schematic configuration of the position data processing unit 4 shown in FIG.

  The encoder device 1 according to the present embodiment is a device for detecting the rotation angle (rotation position) of the rotating body, and specifically, for example, a device for detecting the rotation angle of the rotor of the servo motor 2. As shown in FIG. 1, the encoder device 1 includes an encoder 3 and a position data processing unit 4. The position data processing unit 4 is connected to the output side of the encoder 3. The output side of the position data processing unit 4 is connected to a motor control unit (motor driver) 5 that drives and controls the servo motor 2.

  The encoder 3 is an absolute encoder (absolute value encoder). The encoder 3 according to the present embodiment is a magnetic rotary encoder. As shown in FIG. 2, the sensor magnet 8 fixed to the rotating shaft 7 or the like constituting the rotor of the servo motor 2 and the sensor magnet 8 are opposed to each other. A magnetoresistive element 9 and two Hall elements 10 are provided.

  The sensor magnet 8 is a permanent magnet formed in a disk shape. On the surface of the sensor magnet 8 facing the magnetoresistive element 9 and the Hall element 10, one N pole and one S pole are formed in the circumferential direction. The magnetoresistive element 9 is arranged so that the center of the sensor magnet 8 and the center of the magnetoresistive element 9 substantially coincide when viewed from the axial direction of the rotating shaft 7. The magnetoresistive element 9 is formed with a magnetoresistive pattern arranged in a direction substantially orthogonal to each other. The two Hall elements 10 are arranged at positions shifted from each other by 90 ° with respect to the center of the sensor magnet 8 when viewed from the axial direction of the rotating shaft 7.

  Based on the detection result of the magnetoresistive element 9, the encoder 3 generates position data (one rotation data) indicating a rotation angle within a range of one rotation (360 °) of the rotating body (rotor). Data is output toward the position data processing unit 4. Further, the encoder 3 generates multi-rotation data indicating how many times the rotating body is rotating from a predetermined origin position based on the detection result of the Hall element 10, and the multi-rotation data is generated as a position data processing unit. Output toward 4. Furthermore, the encoder 3 outputs error state data indicating an error state of the encoder 3 toward the position data processing unit 4. For example, when position data or the like to be output from the encoder 3 is not output, the encoder 3 outputs error state data. The encoder 3 communicates with the position data processing unit 4 at a constant sampling cycle. That is, the position data processing unit 4 acquires position data, multi-rotation data, and the like at a constant sampling period.

  The position data processing unit 4 is a microprocessor (MPU). The position data processing unit 4 performs a predetermined process on the position data (one rotation data) before processing output from the encoder 3 to generate position data (mechanical angle) after processing. That is, the position data processing unit 4 processes the pre-processing position data output from the encoder 3 to generate post-processing position data. As shown in FIG. 3, the position data processing unit 4 includes a scaling unit 15 to which pre-processing position data is input. The scaling unit 15 scales the pre-processing position data.

  Connected to the output side of the scaling unit 15 is a first-order difference value calculation unit 16 that calculates pre-processing first-order difference value data, which is the difference value of the pre-processing position data after scaling acquired at a certain sampling period. ing. The first-order difference value calculation unit 16 determines the difference between the pre-processing position data acquired this time and the pre-processing position data acquired last time (that is, the pre-processing position data at the time of the current sampling, Pre-process first-floor difference value data is calculated by a backward difference that takes a difference from the pre-process position data at the time of sampling. That is, the first floor difference value calculation unit 16 calculates the first floor difference value data before processing at a constant sampling period. The pre-processing first floor difference value data is data corresponding to the rotational speed of the rotating body at the time of the current sampling. Connected to the output side of the first-order difference value calculation unit 16 is a range conversion unit 17 that performs range conversion on the first-order difference value data before processing.

  On the output side of the range conversion unit 17, a limiter unit 18 is connected for evaluating whether or not the pre-processing first-order difference value data is within a predetermined reference range (first reference range). An OR circuit 19 is connected to the output side of the limiter unit 18. The limiter unit 18 outputs a signal to the OR circuit 19 when the pre-processing first-order difference value data is out of the first reference range, and the pre-processing first-order difference value data is within the first reference range. In this case, no signal is output to the OR circuit 19. The upper limit value of the first reference range is a value that is higher than the maximum rotation speed (maximum rotation speed) of the servo motor 2, and is a value that is set empirically. Further, the upper limit value and the lower limit value of the first reference range are values that cannot occur during normal rotation of the servo motor 2.

  In addition, a second-order difference value calculation unit 20 that calculates second-order difference value data that is a difference value of the first-order difference value data before processing is connected to the output side of the range conversion unit 17 for each sampling period. The second-order difference value calculation unit 20 calculates the difference between the pre-processing first-order difference value data calculated this time and the pre-processing first-order difference value data calculated last time (that is, before the processing at the time of the current sampling) in the sampling period. The second-order difference value data is calculated by a backward difference that takes the difference between the first-order difference value data and the first-step difference value data before processing at the time of the previous sampling. That is, the second-order difference value calculation unit 20 calculates second-order difference value data at a constant sampling period. The second-order difference value data is data corresponding to the acceleration of the rotating body during the current sampling.

  A limiter unit 21 for evaluating whether or not the second-order difference value data is within a predetermined reference range (second reference range) is connected to the output side of the second-order difference value calculation unit 20. An OR circuit 19 is connected to the output side of the limiter unit 21. The limiter unit 21 outputs a signal to the logical sum circuit 19 when the second-order difference value data is out of the second reference range, and when the second-order difference value data is within the second reference range, No signal is output to the sum circuit 19. The upper limit value of the second reference range is a value that is higher than the maximum acceleration of the servo motor 2 and is an empirically set value. Further, the upper limit value and the lower limit value of the second reference range are values that cannot occur during normal rotation of the servo motor 2.

  A logical sum circuit 22 is connected to the output side of the logical sum circuit 19. The OR circuit 19 outputs a signal toward the OR circuit 22 when a signal is input from at least one of the limiter unit 18 and the limiter unit 21.

  Further, on the output side of the scaling unit 15, the difference value between the pre-processing position data after the scaling acquired this time and the post-processing position data generated the last time (that is, before the processing at the time of the current sampling) A first-order difference value calculation unit 25 that calculates first-order difference value data before and after processing, which is a difference value between position data and post-processing position data generated at the time of previous sampling, is connected. The first floor difference value calculation unit 25 calculates the first floor difference value data before and after processing at a constant sampling period. Connected to the output side of the first-order difference value calculation unit 25 is a range conversion unit 26 that performs range conversion on the first-order difference value data before and after processing.

  Connected to the output side of the range conversion unit 26 is a limiter unit 27 for evaluating whether or not the first-order difference value data before and after the range conversion exceeds a predetermined reference value (third reference value). Yes. A data switching unit 28 is connected to the output side of the range converter 26.

  An OR circuit 22 is connected to the output side of the limiter unit 27. The limiter unit 27 outputs a signal to the OR circuit 22 when the first-order difference value data before and after processing exceeds the third reference value, and the first-order difference value data before and after processing is within the third reference value. In this case, no signal is output to the OR circuit 22. The third reference value is a value higher than the maximum rotation speed (maximum rotation speed) of the servo motor 2, and is a value set empirically. The third reference value is a value that cannot be generated when the servo motor 2 rotates normally. The third reference value of the present embodiment is the same value as the upper limit value of the first reference range, but the third reference value may be different from the upper limit value of the first reference range.

  A data switching unit 28 and a low-pass filter 29 are connected to the output side of the limiter unit 27. The limiter unit 27 outputs the data of the third reference value as the first-order difference value data before and after processing to the data switching unit 28 and the low-pass filter 29 when the first-order difference value data before and after processing exceeds the third reference value. To do. That is, the limiter unit 27 performs a saturation process, and outputs the first-order difference value data before and after the saturation process to the data switching unit 28 and the low-pass filter 29. The limiter unit 27 does not output the first-order difference value data before and after processing to the data switching unit 28 and the low-pass filter 29 when the first-order difference value data before and after processing is within the third reference value. A data switching unit 28 is connected to the output side of the low-pass filter 29. The low-pass filter 29 removes high-frequency components from the first-order difference value data before and after the saturation process output from the limiter unit 27, and outputs first-order difference value data before and after the processing after removal of the high-frequency components.

  A data switching unit 30 is connected to the output side of the data switching unit 28. The data switching unit 28 is a first-order difference value data before and after processing output from the range conversion unit 26, a first-order difference value data before and after processing output from the limiter unit 27, and a first-order difference before and after processing output from the low-pass filter 29. One of the difference value data before and after the process is selected from the difference value data according to a predetermined condition, and is output to the data switching unit 30.

  A logical sum circuit 31 is connected to the output side of the logical sum circuit 22. The logical sum circuit 22 outputs a signal toward the logical sum circuit 31 when a signal is input from at least one of the logical sum circuit 19 and the limiter unit 27. Further, when a signal is input to the OR circuit 22 from at least one of the OR circuit 19 and the limiter unit 27, a data abnormality signal is output from the position data processing unit 4.

  Error state data output from the encoder 3 can be input to the logical sum circuit 31. The OR circuit 31 outputs a signal toward the data switching unit 30 when at least one of the signal from the OR circuit 22 and the error state data from the encoder 3 is input.

  When a signal is input from the OR circuit 31, the data switching unit 30 calculates the average value of the first-order differential value data before and after the predetermined number of processes calculated up to the previous sampling time (that is, calculated up to the previous sampling time). The moving average value of the first floor difference value data before and after the process) is estimated as the first floor difference value data before and after the current process, and this average value is output as the first floor difference value data before and after the current process. That is, in this embodiment, the situation in which the first floor differential value data before processing is out of the first reference range, the situation in which the second floor differential value data is out of the second reference range, and the first floor differential value data before and after processing are the third The difference between the first floor before and after processing calculated up to the previous sampling time when at least one of the situation exceeding the reference value and the situation where the error status data is output from the encoder 3 has occurred. The moving average value of the value data becomes the first-order difference value data before and after the current process output from the data switching unit 30.

  In addition, when no signal is input from the OR circuit 31, the data switching unit 30 outputs the pre- and post-processing first-order difference value data output from the data switching unit 28 as the current pre- and post-processing first-order difference value data. Here, when the first-order difference data before and after processing is within the third reference value, no signal is input from the OR circuit 31 to the data switching unit 30. Further, the limiter unit 27 does not output the first-order difference value data before and after the saturation process to the data switching unit 28 and the low-pass filter 29 when the first-order difference value data before and after the process is within the third reference value. Therefore, when no signal is input from the logical sum circuit 31 to the data switching unit 30, the data switching unit 30 uses the first-order difference value data before and after the current process output from the range conversion unit 26 as the first-order difference value before and after the current process. Output as data. That is, in this embodiment, when no signal is input from the OR circuit 31 to the data switching unit 30, the first-stage difference value data before and after the process calculated at the time of the current sampling is output from the data switching unit 30 before and after the current process 1. It becomes floor difference value data.

  The first-stage difference value data before and after the current process output from the data switching unit 30 is output from the position data processing unit 4 as mechanical angle difference value data.

  A cumulative addition unit 35 is connected to the output side of the data switching unit 30. The cumulative addition unit 35 calculates the sum of the first-order difference value data before and after the current process output from the data switching unit 30 and the first-order difference value data before and after the process calculated up to the previous sampling. Output value data. The total value data is output from the position data processing unit 4 as mechanical angle integrated value data.

  In addition, a cumulative addition unit 36 is connected to the output side of the data switching unit 30. A range converter 37 is connected to the output side of the cumulative adder 36. The cumulative addition unit 36 calculates the sum of the first-order difference value data before and after the current process output from the data switching unit 30 and the first-order difference value data before and after the process calculated up to the previous sampling, The conversion unit 37 performs range conversion on this sum. The data output from the range conversion unit 37 is post-processing position data generated by the position data processing unit 4. The post-processing position data is output from the position data processing unit 4 as mechanical angle data. A first-order difference value calculation unit 25 is connected to the output side of the range conversion unit 37, and post-processing position data generated by the range conversion unit 37 is input to the first-order difference value calculation unit 25.

  In addition, a moving average value calculation unit 38 that calculates a moving average value of the first-order difference value data before and after processing is connected to the output side of the data switching unit 30. The moving average value calculation unit 38 according to the present embodiment calculates the moving average value of the first-order difference value data before and after the process by the simple moving average. The moving average value calculated by the moving average value calculation unit 38 is input to the data switching unit 30. When a signal is input from the logical sum circuit 31 to the data switching unit 30, the moving average value calculated by the moving average value calculating unit 38 at the previous sampling is converted from the data switching unit 30 as the first-order difference value data before and after the current process. Is output.

  An electrical angle conversion unit 39 that converts a mechanical angle into an electrical angle is connected to the output side of the range conversion unit 37 and the output side of the moving average value calculation unit 38. The electrical angle conversion unit 39 generates and outputs normalized electrical angle data and electrical angle data. These data generated by the electrical angle conversion unit 39 are output from the position data processing unit 4.

(Pre-processing position data error detection method)
FIG. 4 is a flowchart for explaining a method for detecting an abnormality in position data before processing in the position data processing unit 4 shown in FIG.

  The pre-processing position data input to the position data processing unit 4 becomes erroneous pre-processing position data due to a bit error due to electromagnetic noise or the like when it is sent from the encoder 3 to the position data processing unit 4. There is. That is, the pre-processing position data input to the position data processing unit 4 may be abnormal due to the influence of electromagnetic noise or the like when being sent from the encoder 3 to the position data processing unit 4. Moreover, since the encoder 3 of this embodiment is a magnetic encoder, it is easily affected by electromagnetic noise, and erroneous pre-processing position data with superimposed noise may be output from the encoder 3. That is, an abnormality may occur in the pre-processing position data input to the position data processing unit 4 inside the encoder 3.

  The position data processing unit 4 of this embodiment detects an abnormality in the position data before processing input to the position data processing unit 4 as follows. That is, when the position data processing unit 4 acquires the pre-processing position data at a constant sampling period, the first-floor difference value calculation unit 16 calculates the pre-processing first-floor difference value data (step S1), and the second-floor difference value. The calculation unit 20 calculates second-order difference value data (step S2). Thereafter, the position data processing unit 4 uses the limiter unit 18 to evaluate whether or not the pre-processing first-order difference value data is within the first reference range (step S3), and uses the limiter unit 21 to It is evaluated whether or not the floor difference value data is within the second reference range (step S4).

  When the first floor differential value data before processing is in the first reference range and the second floor differential value data is in the second reference range (“Yes” in step S3 and “Yes” in step S4) In this case, since it is estimated that the pre-processing position data within the assumption is acquired at the time of the current sampling, the position data processing unit 4 determines that the position data before the processing at the time of the current sampling is normal. Determine (step S5).

  On the other hand, when the pre-processing first floor difference value data is out of the first reference range (if “No” in step S3), or when the second floor difference value data is out of the second reference range (step In the case of “No” in S4), since it is estimated that unexpected pre-processing position data is acquired at the time of the current sampling, the position data processing unit 4 performs the pre-processing position data at the time of the current sampling. Is determined to be abnormal (step S6). That is, the position data processing unit 4 performs the sampling at this time when the pre-processing first-order difference value data is out of the first reference range or when the second-order difference value data is out of the second reference range. It is detected that the pre-processing position data is abnormal.

  Step S1 of the present embodiment is a first-order difference value calculation step for calculating pre-processing first-order difference value data, which is a difference value of pre-processing position data acquired every predetermined sampling period, and step S2 is a sampling period. This is a second-order difference value calculation step for calculating second-order difference value data, which is the difference value of the first-order difference value data before processing. Step S3 is a first evaluation step for evaluating whether or not the first floor difference value data before processing is within a predetermined first reference range, and step S4 is a second step where the second floor difference value data is a predetermined second value. It is a 2nd evaluation step which evaluates whether it is in a standard range.

(Method for generating post-processing position data)
FIG. 5 is a flowchart for explaining a method of generating post-processing position data in the position data processing unit 4 shown in FIG.

  The position data processing unit 4 generates post-process position data as follows. That is, when the position data processing unit 4 acquires the pre-processing position data at a constant sampling cycle, the first-floor difference value calculation unit 25 calculates first-floor first-order difference value data before and after processing (step S11). Thereafter, the position data processing unit 4 uses the limiter unit 27 to evaluate whether or not the first-order difference data before and after processing is within the third reference value (step S12).

  If the difference data of the first floor before and after the processing is within the third reference value (if “Yes” in step S12), the position data processing unit 4 performs the above-described abnormality detection processing of the position data before processing. It is determined whether or not the pre-processing position data at the time of sampling is determined to be normal (step S13). When it is determined that the first floor difference value data before and after the processing is within the third reference value and the pre-processing position data at the time of this sampling is normal (in the case of “Yes” in step S13), Since no signal is input from the OR circuit 31 to the data switching unit 30, the first-order difference value data before and after processing calculated at the time of the current sampling is output from the data switching unit 30 as first-order difference value data before and after the current processing. (Step S14).

  On the other hand, when the difference data before and after the processing exceeds the third reference value (when “No” in step S12), or when it is determined that the pre-processing position data at the time of the current sampling is abnormal ( In the case of “No” in step S13), since a signal is input from the OR circuit 31 to the data switching unit 30, the moving average value of the first-order difference value data before and after processing calculated up to the previous sampling time Is estimated as first-order difference value data before and after the current process, and this average value is output from the data switching unit 30 as first-order difference value data before and after the current process (step S15).

  Thereafter, the cumulative addition unit 36 and the range conversion unit 37 perform cumulative addition and the like to generate post-processing position data (step S16). In step S16, when the first-order difference value data before and after the process calculated at the time of the current sampling is output from the data switching unit 30 as the first-order difference value data before and after the current process, the cumulative addition unit 36 and the range conversion unit 37 are output. Generates the post-processing position data of this time using the first-order differential value data before and after processing calculated at the time of sampling. Specifically, the cumulative addition unit 36 and the range conversion unit 37 calculate the sum of the first-order difference value data before and after processing calculated during the current sampling and the first-order difference value data before and after processing calculated up to the previous time. The calculated post-processing position data is generated. In this case, the position data processing unit 4 uses the post-processing position data calculated based on the first-order difference value data before and after the processing based on the actual pre-processing position data input from the encoder 3 at the time of the current sampling. The motor control unit 5 controls the servo motor 2 based on the post-processing position data.

  Further, in step S16, when the moving average value of the first floor difference value data before and after the process calculated up to the previous sampling is output from the data switching unit 30 as the first floor difference value data before and after the current process, The cumulative addition unit 36 and the range conversion unit 37 generate current post-processing position data using the moving average value of the first-order difference value data before and after processing calculated up to the previous sampling. Specifically, the cumulative addition unit 36 and the range conversion unit 37 calculate the moving average value of the first-order difference value data before and after processing calculated up to the time of the previous sampling and the first-order value difference before and after processing calculated up to the previous time. The sum total with the value data is calculated, and the current post-processing position data is generated. In this case, the position data processing unit 4 is based on the moving average value of the first-order difference value data before and after the process calculated by the previous sampling (that is, the estimated value of the first-order difference value data before and after the process). The calculated post-processing position data is output to the motor control unit 5, and the motor control unit 5 controls the servo motor 2 based on the post-processing position data.

(Main effects of this form)
As described above, in this embodiment, the position data processing unit 4 calculates the first floor difference value data and the second floor difference value data before processing, and the first floor difference value data before processing deviates from the first reference range. Or when the second-order difference value data is out of the second reference range, it is determined that the pre-processing position data is abnormal, and the pre-processing first-order difference that is the difference value of the pre-processing position data In addition to the value data, an abnormality in the pre-processing position data is detected by using the second-order difference value data that is the difference value of the pre-processing first-order difference value data. That is, in this embodiment, not only when there is a sudden change in the rotation speed of the rotating body calculated based on the pre-processing position data, but also a sudden change in the acceleration of the rotating body calculated based on the pre-processing position data. Even when there is, an abnormality in the position data before processing is detected. For this reason, in this embodiment, it is possible to improve the accuracy of detecting the abnormality of the pre-processing position data as compared with the case of detecting the pre-processing position data abnormality using only the pre-processing first-order difference value data.

  In this embodiment, when the position data processing unit 4 determines that the pre-processing position data at the time of the current sampling is normal, the position data processing unit 4 uses the first-order difference value data before and after the processing calculated at the time of the current sampling, Position data is generated. Therefore, in this embodiment, the servo motor 2 can be controlled by the post-processing position data generated based on the normal pre-processing position data, and the rotation characteristics of the servo motor 2 can be prevented from being adversely affected. Is possible. Further, in this embodiment, when the position data processing unit 4 determines that the pre-processing position data at the time of the current sampling is abnormal, the moving average value of the first-order difference value data before and after the processing calculated up to the previous sampling time Is used to generate the post-processing position data. Therefore, in this embodiment, it is possible to prevent the occurrence of a phenomenon that the behavior of the servo motor 2 becomes unstable due to post-processing position data generated based on abnormal pre-processing position data.

  In the present embodiment, if the first and second floor difference value data calculated by the first floor difference value calculation unit 25 exceeds the third reference value, the position data processing unit 4 performs the process calculated up to the previous sampling time. The post-processing position data is generated using the moving average value of the front and back first-order difference value data. When the first and second floor difference value data before and after processing calculated by the first floor difference value calculation unit 25 exceeds the third reference value, it is assumed that the pre-processing position data acquired at the time of this sampling is abnormal. In addition, it is assumed that the post-processing position data generated last time is an estimated value generated using the moving average value of the first-order difference data before and after the processing. It is assumed that the value is far from the position data. Calculated based on the estimated post-processing position data at the previous sampling when the estimated post-processing position data at the previous sampling is significantly different from the normal pre-processing position data. If the current post-processing position data is generated based on the first-stage difference value data before and after the current processing, the generated current post-processing position data may become a value far from the correct post-processing position data. There is.

  However, in this embodiment, when the first and second floor difference value data calculated by the first floor difference value calculation unit 25 exceeds the third reference value, the position data processing unit 4 calculates the previous sampling time. Since the post-processing position data is generated using the moving average value of the first-order difference data before and after the processing, the estimated post-processing position data is far from the correct post-processing position data. It becomes possible to prevent. That is, in this embodiment, it is possible to suppress the difference between the estimated post-processing position data this time and the correct post-processing position data. Accordingly, the estimated post-processing position data can be made to follow the normal pre-processing position data, and the estimated post-processing position data can be prevented from greatly deviating from the normal pre-processing position data. It becomes possible. As a result, in this embodiment, it is possible to stabilize the behavior of the servo motor 2 that is controlled based on the post-processing position data.

(Other embodiments)
The above-described embodiment is an example of a preferred embodiment of the present invention, but is not limited to this, and various modifications can be made without departing from the scope of the present invention.

  In the above-described form, when the first floor difference value data before and after processing exceeds the third reference value, the moving average value of the first floor difference value data before and after processing calculated up to the previous sampling is used as the first floor difference before and after the current processing. It is estimated as value data, and this average value is output from the data switching unit 30. In addition to this, for example, when the first-order difference value data before and after the processing exceeds the third reference value, the first-order difference value data before and after the processing calculated at the time of the current sampling is used as the first-order difference value data before and after the current processing. You may output from the switch part 30. FIG.

  In the above-described form, as shown in the flowchart of FIG. 4, the position data processing unit 4 evaluates whether or not the pre-processing first-order difference value data is within the first reference range, and then the second-order difference value data is It is evaluated whether or not it is within the second reference range. In addition to this, for example, the position data processing unit 4 evaluates whether or not the second-order difference value data is within the second reference range, and then determines whether or not the first-order difference value data before processing is within the first reference range. Evaluation of whether or not the first-stage difference value data before processing is within the first reference range and evaluation of whether or not the second-order difference value data is within the second reference range May be performed simultaneously.

  In the above-described form, the position data processing unit 4 determines that the pre-processing position data at the time of this sampling is normal after evaluating whether or not the first-order difference data before and after the processing is within the third reference value. It is determined whether or not it has been done. In addition to this, for example, after the position data processing unit 4 determines whether or not the pre-processing position data at the time of the current sampling is determined to be normal, the first-order difference data before and after the processing is within the third reference value. It is possible to evaluate whether or not the first-order difference data before and after processing is within the third reference value, and determine that the pre-processing position data at the time of sampling is normal. The determination of whether or not it has been performed may be performed simultaneously.

  In the embodiment described above, the encoder 3 is an absolute encoder, but the encoder 3 may be an incremental encoder. In the embodiment described above, the encoder 3 is a magnetic encoder. The encoder 3 is, for example, a slit plate in which a light emitting element and a light receiving element and a slit that transmits light from the light emitting element to the light receiving element are formed. May be a photoelectric encoder. Further, the encoder 3 may be an encoder other than a magnetic type or a photoelectric type. Moreover, in the form mentioned above, although the encoder 3 is a rotary encoder, the encoder 3 may be a linear encoder.

DESCRIPTION OF SYMBOLS 1 Encoder apparatus 3 Encoder 4 Position data processing part S1 1st floor difference value calculation step S2 2nd floor difference value calculation step S3 1st evaluation step S4 2nd evaluation step

Claims (5)

An encoder, and a position data processing unit that processes post-processing position data output by the encoder and generates post-processing position data;
The position data processing unit
Calculating pre-processing first-order difference value data that is a difference value of the pre-processing position data acquired at each predetermined sampling period;
Calculating second-order difference value data that is a difference value of the first-order difference value data before processing for each sampling period;
When the pre-processing first floor difference value data is within a predetermined first reference range and the second floor difference value data is within a predetermined second reference range, the pre-processing position data is normal. Judgment,
The position data before processing is abnormal when the first floor difference value data before processing is out of the first reference range and / or when the second floor difference value data is out of the second reference range. It is determined that
Furthermore, in the sampling period, before and after the first-stage difference value data before and after processing, which is a difference value between the post-processing position data generated last time and the pre-processing position data acquired this time,
When it is determined that the pre-processing position data is normal, the post-processing position data of this time is generated using the first-order difference value data before and after the processing calculated this time in the sampling period,
If it is determined that the pre-processing position data is abnormal, the average value of the predetermined number of first- and second-stage difference value data calculated up to the previous time in the sampling period is calculated as the first-stage difference value data before and after the current process. The encoder apparatus is characterized in that the post-processing position data of this time is generated using the average value . The position data processing unit
When it is determined that the pre-processing position data is normal, in the sampling period, the first-order difference value data before and after the processing calculated this time and the first-order difference value data before and after the processing calculated up to the previous time Calculate the sum, generate the post-processing position data for this time,
If it is determined that the pre-processing position data is abnormal, in the sampling period, the average value of the predetermined number of first- and post-processing first-order difference data calculated up to the previous time and the pre-processing and post-processing 1 calculated up to the previous time It calculates the sum of the Kaichi differential value data, the encoder apparatus according to claim 1, wherein generating a current of the processed position data. The position data processing unit
If it is determined that the pre-processing position data is normal and the first- and second-stage difference value data before and after the processing is within a predetermined third reference value, the first floor before and after the processing calculated this time in the sampling period Using the difference value data, generate the post-processing position data of this time,
Even if it is determined that the pre-processing position data is normal, if the first-order differential value data before and after the processing exceeds the third reference value, the predetermined number of the calculated pieces up to the previous time in the sampling period The average value of the first floor difference value data before and after the processing is estimated as the first floor difference value data before and after the current processing, and the current post-processing position data is generated using the average value. The encoder device according to 1 or 2 . The position data processing unit
If it is determined that the pre-processing position data is normal and the pre- processing and post-processing first-order difference value data is within the third reference value, the currently calculated pre-processing and post-processing first-order difference is calculated in the sampling period. Calculating the sum of the value data and the first-order difference value data before and after the processing calculated up to the previous time, and generating the post-processing position data of this time,
Even if it is determined that the pre-processing position data is normal, if the first-order differential value data before and after the processing exceeds the third reference value, the predetermined number of the calculated pieces up to the previous time in the sampling period Calculating the sum of the average value of the first floor difference value data before and after processing and the first floor difference value data before and after processing calculated up to the previous time, and generating the post-processing position data of this time The encoder device according to claim 3 . A position data abnormality detection method in an encoder device comprising an encoder and a position data processing unit that processes post-processing position data output by the encoder and generates post-processing position data,
A first-order difference value calculating step of calculating pre-processing first-order difference value data, which is a difference value of the pre-processing position data acquired every predetermined sampling period;
A second-order difference value calculating step of calculating second-order difference value data that is a difference value of the pre-processing first-order difference value data for each sampling period;
A first evaluation step for evaluating whether or not the pre-processing first floor difference value data is within a predetermined first reference range;
A second evaluation step for evaluating whether the second-order difference value data is within a predetermined second reference range;
When the first floor differential value data before processing is within the first reference range in the first evaluation step, and the second floor differential value data is within the second reference range in the second evaluation step, Detecting that the pre-processing position data is normal;
When the first- stage differential value data before processing is out of the first reference range in the first evaluation step and / or, the second-order difference value data is out of the second reference range in the second evaluation step. And detecting that the pre-processing position data is abnormal ,
Further, in the sampling period, calculating first-order differential value data before and after processing that is a difference value between the post-processing position data generated last time and the pre-processing position data acquired this time;
When it is detected that the pre-processing position data is normal, the post-processing position data of this time is generated using the first-order differential value data before and after the processing calculated this time in the sampling period, When it is detected that the pre-processing position data is abnormal, the average value of the predetermined number of first- and second-stage difference value data calculated up to the previous time in the sampling period is calculated as the first-stage difference before and after the current process. A position data anomaly detection method comprising: estimating the value data and generating the post-processing position data of this time using the average value .

Images