JP6111642B2 - Control device, control program, and control method - Google Patents

Description

  The present invention relates to a control device, a control program thereof, and a control method thereof related to a spindle / slave interlocking control application.

  A control device such as a programmable logic controller (hereinafter referred to as “PLC”) is used for controlling machines and equipment. Since such a control device performs a control calculation at a predetermined cycle, a control delay time (dead time) may be a problem.

  As an example in which such a control delay time becomes a problem, there is a control application interlocking with a main shaft and a slave shaft. As an example of this spindle / slave-linked control application, processing on an object using measurement information obtained by measuring the physical displacement (rotation speed or displacement) of the object with a measuring device (spindle). This is an application that controls (slave shaft). In such a control application, there is a problem that a delay occurs until information measured on the main shaft is processed and transmitted to the slave shaft. Therefore, a method for reducing the influence of such a delay has been proposed as follows.

  Japanese Patent Application Laid-Open No. 08-126375 (Patent Document 1) discloses a synchronous control device in which a sub shaft motor can follow a main shaft motor without delay. More specifically, in a device for synchronously driving the main shaft motor and the sub shaft motor, a foreseeing controller that is controlled so that the rotational position of the motor predicted from the dynamic characteristic model of the sub shaft motor matches the future target position command. And a means for calculating a predicted value of the future position of the spindle motor, and a means for obtaining a future target position command of the dependent axis motor as a function of the predicted future position of the spindle motor and inputting it to the foreseeing controller. Is disclosed.

  Japanese Patent Application Laid-Open No. 09-289788 (Patent Document 2) realizes highly accurate synchronization control, and there is little deterioration in synchronization accuracy even when there is a delay in position command input or spindle position detection. A synchronous control device is disclosed. More specifically, a synchronous control device that drives the dependent shaft motor in synchronization with the main shaft motor, including the prediction device 1, the main shaft device 2, and the dependent shaft device 3, is disclosed. The prediction device 1 outputs a spindle position command increment value input to a plurality of past points up to the present, and a spindle position command increment value input before M-1 sampling among the stored values. Means for storing the spindle position increment values inputted to a plurality of past points up to the present, a dynamic characteristic model of the spindle device including d sampling delay, and the stored spindle position command increment value And a calculator 9 for obtaining predicted values of a plurality of spindle position increment values up to several sampling destinations based on the spindle position increment value, and a plurality of dependent axis future positions from the obtained predicted values of the plurality of spindle position increment values. A configuration including a converter 10 for obtaining a command is disclosed.

  Furthermore, Japanese Patent Laid-Open No. 10-174478 (Patent Document 3) discloses a synchronous control device that does not deteriorate the synchronization accuracy even when the dynamic characteristics of the main shaft are different between forward rotation and reverse rotation. Specifically, it is a synchronous control device that drives a subordinate motor in synchronization with a main shaft motor, and includes a signal obtained by multiplying an increment value between sampling periods of the main shaft position or a predicted value thereof by a multiplier K1, and a main shaft position command increment. In a synchronous control device provided with a fine adjustment device 19 for inputting a signal obtained by adding a signal multiplied by a multiplier K2 to a control device for a dependent axis motor as a fine adjustment signal for synchronization deviation, two types of multipliers K1 are stored, and the spindle A configuration having means for switching depending on whether the position increment value is positive or negative is disclosed.

Japanese Patent Laid-Open No. 08-126375 JP 09-289788 A Japanese Patent Laid-Open No. 10-174478

  In the above-described Patent Documents 1 to 3, it is necessary to adjust a model, a parameter, and the like according to a target control system. When actually applied, adjustment takes time.

  An object of the present invention is to provide a novel control device, control program, and control method thereof that can reduce the influence due to a control delay time that occurs in a control application interlocked with a main shaft and a slave shaft.

  A control device according to an aspect of the present invention includes an input unit that acquires measurement information indicating physical displacement of an object from the measurement device, a drive unit that drives a processing device that performs processing on the object, and measurement information. And a control unit for controlling the drive unit based on the control unit. The control unit corrects the measurement information before the smoothing process by comparing the measurement information with the smoothing means for performing the smoothing process on the measurement information and the result of the smoothing process calculated in advance. Correction means.

  Preferably, the correcting means includes means for calculating a difference between the measurement information and the result of the smoothing process calculated in advance.

  Preferably, the correction means calculates a difference between the means for shifting the phase to the delay side with respect to the result of the smoothing process calculated in advance, and the measurement information and the result of the smoothing process with the phase shifted. Including means.

  Preferably, the measuring device measures the position at each time point related to the movement of the object as a physical displacement, and the smoothing means smoothes the moving speed that is a temporal change amount of the position of the object.

  Preferably, at least one of the input unit and the drive unit exchanges information through a network.

Preferably, a control part performs a calculation process with a predetermined period.
Preferably, the control unit further includes a calculation unit that calculates a command value to be given to the drive unit based on the result of the smoothing process, and the calculation unit estimates the position of the object from the result of the smoothing process.

  According to another aspect of the present invention, an input unit that acquires measurement information indicating physical displacement of an object from a measurement device, a drive unit that drives a processing device that performs processing on the object, and a control unit, There is provided a control program to be executed by a computer comprising: The control program compares the measurement information with the step of performing the smoothing process on the measurement information to the computer by comparing the result of the smoothing process calculated in advance with the measurement information. And a step of correcting.

  According to still another aspect of the present invention, an input unit that acquires measurement information indicating physical displacement of an object from a measurement device, a drive unit that drives a processing device that performs processing on the object, and a control unit And a control method executed by a computer. The control method includes a step of performing the smoothing process on the measurement information and a step of correcting the measurement information before the smoothing process by comparing the measurement result with the result of the smoothing process calculated in advance. Including.

  ADVANTAGE OF THE INVENTION According to this invention, the influence by the control delay time which arises in the control application of a spindle / slave interlock can be reduced.

It is a schematic diagram which shows the structure of the control system which concerns on this Embodiment. It is a schematic diagram for demonstrating a spindle / slave interlock in the control system which concerns on this Embodiment. It is a figure for demonstrating the influence of the delay time in the control system which concerns on this Embodiment. It is a figure for demonstrating the influence of the delay time in the control system which concerns on this Embodiment. It is a schematic diagram which shows the control structure mounted in the control apparatus which concerns on this Embodiment. It is a flowchart which shows the process sequence in the control system which concerns on this Embodiment. It is a figure which shows an example which verified the improvement effect by the control system which concerns on this Embodiment by simulation. It is a schematic diagram which shows the control structure mounted in the control apparatus which concerns on a 1st modification. It is a figure which shows an example which verified the improvement effect by the control apparatus which concerns on a 1st modification by simulation. It is a figure which shows an example which verified the improvement effect by the control apparatus which concerns on a 1st modification by simulation. It is a schematic diagram which shows the structure of the control system which concerns on a 2nd modification. It is a flowchart which shows the process sequence in the control system which concerns on a 2nd modification. It is a schematic diagram which shows the whole control structure mounted in the control system which concerns on a 2nd modification. It is a figure which shows an example of the effect of the target position prediction relevant to the control system which concerns on a 2nd modification. It is a figure which shows an example of the effect of another target position prediction relevant to the control system which concerns on a 2nd modification. It is a figure which shows an example of the effect of the target position prediction by the control system which concerns on another form of a 2nd modification. It is a schematic diagram which shows the whole control structure mounted in the control system which concerns on a 3rd modification. It is a figure for demonstrating the production method of the hint data contained in the input data input into the control apparatus which concerns on a 3rd modification. It is a figure which shows an example of the effect of the future position prediction by the control apparatus which concerns on a 3rd modification. It is a schematic diagram which shows the control structure mounted in the control apparatus which concerns on a 4th modification. It is a figure for demonstrating the principle for compensating for the time delay in the control structure shown in FIG.

  Embodiments of the present invention will be described in detail with reference to the drawings. In addition, about the same or equivalent part in a figure, the same code | symbol is attached | subjected and the description is not repeated.

<A. System configuration>
As an embodiment of the present invention, a spindle / slave-linked control application that measures a moving object (hereinafter referred to as “work”) and performs processing at a predetermined position will be considered.

  FIG. 1 is a schematic diagram showing a configuration of a control system 100 according to the present embodiment. Referring to FIG. 1, control system 100 according to the present embodiment includes transport motors 112 and 114 for transporting work W such as an iron plate, and a rotary cutter for cutting work W.

  The movement amount (or conveyance speed) of the workpiece W is measured by a measuring roll associated with the movement of the workpiece W. As this measuring roll, an encoder 140 arranged so that its rotating shaft is in contact with the back surface of the workpiece W is used. That is, the amount of movement of the workpiece W is calculated based on the detection signal (pulse signal) from the encoder 140, and the workpiece W is cut at a predetermined interval.

  The workpiece W is cut by a pair of rotary cutters 102 and 104 arranged in a direction orthogonal to the conveying direction. More specifically, when the rotary cutter 102 is driven by the servo motor 110, the rotary cutters 102 and 104 rotate synchronously, and the workpiece W is cut with a blade formed on a part of the circumference thereof. The servo motor 110 is driven to rotate by being supplied with a signal (typically a pulse signal) related to the drive from the servo driver 40.

  In the example shown in FIG. 1, the encoder 140 is also referred to as a main shaft, and the servo motor 110 that is driven by receiving information from the main shaft is also referred to as a slave shaft.

  The measurement of the movement amount of the workpiece W and the cutting of the workpiece W are controlled by the control device 1. Next, details of the control device 1 will be described.

<B. Configuration of control device>
The control device 1 shown in FIG. 1 is typically realized by a programmable logic controller (hereinafter referred to as “PLC”) or the like. Of course, you may mount not using PLC but using a personal computer (PC) and various arithmetic processing units.

  The control apparatus 1 communicates with a field device such as a CPU (Central Processing Unit) unit 10 that executes main arithmetic processing, a pulse counter unit 20 that detects and counts pulse signals output from the encoder 140, and a servo driver 40. Fieldbus interface (I / F) unit 30 for performing

  The CPU unit 10 includes, as main components, a microprocessor, a volatile memory such as a RAM (Random Access Memory), and a nonvolatile memory such as an HDD (Hard Disk Drive) and a ROM (Read Only Memory). Typically, in the control device 1, processing described later is realized by the CPU unit 10 executing a program. However, all or part of the processing realized by the program may be realized by dedicated hardware.

  As the field network, various industrial Ethernets (registered trademark) such as EtherCAT (registered trademark), Profinet IRT, MECHATRLINK (registered trademark) -III, Powerlink, SERCOS (registered trademark) -III, and CIP Motion can be used. Alternatively, a proprietary network such as DeviceNet or CompoNet / IP (registered trademark) may be used.

  The control device 1 may include an input / output unit and a special unit (not shown). These units are configured to exchange data with each other via a PLC system bus (not shown).

  In the configuration example shown in FIG. 1, the information may be transmitted to the encoder 140 through a network such as a fieldbus. That is, at least one of the pulse counter unit 20 as an input unit and the servo driver 40 as a drive unit may exchange information through a network.

  The pulse counter unit 20 illustrated in FIG. 1 corresponds to an input unit that acquires measurement information indicating a movement amount (or a conveyance speed) that is a physical displacement of an object (work W) from an encoder 140 that is a measurement device. The fieldbus interface unit 30 and the servo driver 40 correspond to a drive unit that drives the rotary cutters 102 and 104 and the servo motor 110 that are processing devices for processing an object (work W). The CPU unit 10 corresponds to a control unit that controls the drive unit based on measurement information (that is, a pulse signal from the encoder 140).

  The control device 1 according to the present embodiment can further improve the cutting accuracy of the workpiece W by employing processing logic as described later. Details of this processing logic will be described below.

<C. About main / slave interlocking>
First, the relationship between the detection value (main shaft position / count value) by the encoder 140 that is the main shaft and the rotational motion of the servo motor 110 that is the sub shaft will be described. The rotary cutter shown in FIG. 1 needs to appropriately control the position of the blade according to the movement of the workpiece W. In other words, by appropriately controlling the rotation angle of the servo motor 110, a kind of cam motion is realized.

  FIG. 2 is a schematic diagram for explaining the interlocking of the main shaft and the slave shaft in the control system 100 according to the present embodiment. As shown in FIG. 2, the slave target angle (the angles of the rotary cutters 102 and 104) is determined according to a predetermined relationship with respect to a change in the spindle position indicating the movement amount of the workpiece W, not linear. Typically, the relationship between the main shaft and the slave shaft is realized by a cam table described later. 2 is defined as a cam table, and the cam table is referred to according to the number of pulses counted by the pulse counter unit 20, so that the rotary cutters 102 and 104 Angles are calculated sequentially.

  In order to make the cut surface with respect to the workpiece W more preferable, both the slave shaft target angle [rad] and the slave shaft target angular velocity [rad / s] which is a differential value thereof are considered. In terms of the slave target angular velocity, the slave target angular velocity is limited to a constant value (typically a value corresponding to the conveyance speed of the workpiece W) from the start position of the cutting operation to the distance p1. Thereafter, the angular velocities of the rotary cutters 102 and 104 are increased so as to maximize at a distance p2 from the starting position of the cutting operation. Thereafter, at the position of the distance p3 from the start position of the cutting operation, the slave target angular velocity is again limited to a constant value (typically a value corresponding to the conveyance speed of the workpiece W). Then, the cutting operation is completed at a position of distance p4 from the starting position of the cutting operation.

Next, the influence of the control delay time in the main shaft / slave shaft linkage will be described.
3 and 4 are diagrams for explaining the influence of the delay time in the control system 100 according to the present embodiment.

  In the control system 100 shown in FIG. 1, the physical change related to the conveyance of the workpiece W is measured by the encoder 140. The encoder 140 measures the position at each time point related to the movement of the workpiece W, which is the object, as a physical displacement. When measurement information by the encoder 140 is transmitted via a field network or the like, a delay time (dead time) due to the transmission is inevitable. Further, the pulse counter unit 20 counts the detection signal (pulse signal) from the encoder 140, and the CPU unit 10 refers to the count value to calculate the slave shaft target angle. Furthermore, a delay time required for the calculated slave shaft target angle to be transmitted to the servo driver 40 via the field bus is also required.

  Due to these delay times (dead time), the timing at which the rotary cutters 102 and 104 as the driven shafts actually start operating is delayed. Specifically, as shown in FIG. 3, since the timing at which the slave target angle is actually effective is delayed with respect to the spindle position, the delay becomes a cut dimension error. That is, originally, the cutting operation should be completed at the position of the distance p4 from the starting position of the cutting operation, but in reality, the cutting operation is completed at the position of the distance p4 '.

  If the conveyance speed of the workpiece W is constant, the delay characteristic shown in FIG. 3 can also be expressed as a characteristic shown in FIG. As shown in FIG. 4, an error in cut dimensions may occur due to a position detection delay or a CPU calculation delay.

  In the present embodiment, by correcting such a delay, the influence of the occurrence of an error in cut dimensions is avoided, and the rotary cutter is controlled with higher accuracy. In particular, in the present embodiment, a correction logic that takes into account a phase delay due to a filter process (smoothing process) executed to suppress a quantization error when calculating the conveyance speed of the workpiece W is employed. More specifically, observer control is introduced in order to correct a phase delay due to filter processing (smoothing processing). Hereinafter, the observer control will be described.

<D. Observer control>
FIG. 5 is a schematic diagram showing a control structure implemented in the control device 1 according to the present embodiment. The control structure shown in FIG. 5 is realized by the CPU unit 10 executing a program. Basically, the CPU unit 10 executes arithmetic processing at a predetermined control cycle (for example, 10 msec). Therefore, the calculation related to the control structure shown in FIG. 5 is also repeatedly executed at a predetermined control cycle.

  Referring to FIG. 5, control device 1 includes, as its control structure, differential element 11, amplification element 12, addition element 13, filter element 14, subtraction element 15, and integration elements 16 and 18. . In the control device 1, the pulse counter unit 20 counts pulse signals from the encoder 140. This count value corresponds to the spindle position of the workpiece W, and this spindle position (count value) is read by the CPU unit 10. The read spindle position (count value) is differentiated with respect to time by the differentiation element 11 (or a difference), whereby the conveyance speed of the workpiece W is calculated. In the control device 1, control is performed based on the conveyance speed of the workpiece W.

  The conveying speed of the workpiece W is smoothed by the filter element 14. The filter element 14 is used to remove the influence of the quantization error caused by the differential element 11 and the noise generated in the pulse signal. That is, the filter element 14 corresponds to a smoothing unit that performs a smoothing process on the pulse signal from the encoder 140 that is measurement information. In the configuration example shown in FIG. 1, the filter element 14 smoothes the conveyance speed, which is a temporal change amount of the position of the workpiece W that is the object.

  The smoothed transport speed output from the filter element 14 corresponds to the estimated transport speed of the workpiece W, and is input to the integrating element 18 and further integrated. The result output from the integration element 18 indicates the position (movement amount) of the workpiece W. The position (movement amount) of the workpiece W output from the integration element 18 is input to the target value calculation unit 19. The target value calculation unit 19 refers to the cam table 19a that defines the characteristics as shown in FIG. 2 and supplies a target value (an angular velocity command value of the servo motor 110 that drives the rotary cutters 102 and 104 or a command value to be given to the servo driver 40). Command angle) is calculated. The calculated target value is output to the servo driver 40. That is, the integration element 18 and the target value calculation unit 19 correspond to calculation means for calculating a command value to be given to the servo motor 110 that is a drive unit based on the result of the smoothing process. And this calculation means estimates the position of the workpiece | work W which is a target object from the result of the smoothing process.

  The filter element 14 is effective in removing the influence of quantization error, noise, etc., but has a certain time constant, so that a delay time occurs between the input and the output. Therefore, the delay time due to the filter element 14 is compensated by observer control. More specifically, the amplification element 12, the addition element 13, the subtraction element 15, and the integration element 16 realize observer control. This observer control is performed by comparing the result of the smoothing process calculated in advance (output of the filter element 14) with the measurement information (value read from the pulse counter unit 20: spindle position). This corresponds to correction means for correcting the spindle position as measurement information before processing.

  That is, the smoothed conveyance speed output from the filter element 14 is input to the integration element 16 and integrated. The result output from the integration element 16 indicates the estimated position (estimated movement amount) of the workpiece W. If there is no delay time as the control system, as an ideal state, the estimated position of the workpiece W coincides with the spindle position read from the pulse counter unit 20. In the present embodiment, since there is a delay time in the filter element 14, there is some difference between the two. By correcting the spindle position based on the difference between the spindle position and the estimated position, a time delay between the two is compensated.

  More specifically, the spindle position read from the pulse counter unit 20 and the estimated position obtained as a result of integration by the integration element 16 are input to the subtraction element 15. Then, the subtraction element 15 calculates the difference between the two and inputs it to the amplification element 12. In this way, the observer control calculates the difference between the measurement information (value read from the pulse counter unit 20: spindle position) and the smoothing processing result (output of the filter element 14) calculated in advance. As a means, a subtraction element 15 is included. The amplification element 12 multiplies the input value by a predetermined correction gain, and then outputs the result to the addition element 13. The adding element 13 adds the output of the amplifying element 12 to the output of the differentiating element 11 to calculate the corrected conveyance speed of the workpiece W.

  By adopting the observer control as described above, it is possible to compensate for the delay time caused by the filter element 14 and perform more accurate position control.

<E. Processing procedure>
Next, a control procedure in the control system 100 shown in FIG. 1 will be described. FIG. 6 is a flowchart showing a processing procedure in the control system 100 according to the present embodiment. Each step shown in FIG. 6 is typically realized by the CPU unit 10 executing a program.

  Referring to FIG. 6, CPU unit 10 determines whether or not work W has reached the cutting position (step S102). Whether or not the workpiece W has reached the cutting position may be determined using a position detection sensor (such as an optical type or a mechanical type) provided in the conveyance path of the workpiece W, or an encoder that is a measuring roller. 140 may be used for the determination.

  If it is determined that the workpiece W has reached the cutting position (YES in step S102), the CPU unit 10 initializes the length of the workpiece W measured by the measuring roll (step S104). Typically, the count value of the detection signal (pulse signal) from the encoder 140 is reset to zero.

  On the other hand, when it is not determined that the workpiece W has reached the cutting position (NO in step S102), the process of step S104 is skipped.

  Subsequently, the CPU unit 10 acquires the count value of the detection signal (pulse signal) from the encoder 140 (step S106). Typically, the pulse counter unit 20 counts the detection signal (pulse signal) from the encoder 140 independently of the control cycle of the CPU unit 10. Then, the CPU unit 10 reads the count value held in the pulse counter unit 20 at a predetermined calculation timing.

  The CPU unit 10 updates the spindle position of the workpiece W (step S108) and calculates the conveyance speed of the workpiece W (step S110). Further, the CPU unit 10 corrects the conveyance speed of the workpiece W using the difference between the estimated position of the workpiece W calculated in the previous control cycle and the spindle position of the workpiece W acquired in the current control cycle (step). S112). Then, the CPU unit 10 performs a filtering process on the corrected conveyance speed of the workpiece W (step S114). When the moving average method is adopted as the filtering process, an average value of the conveyance speed of the workpiece W calculated before that is used.

  Then, the CPU unit 10 calculates the estimated position of the workpiece W from the conveyance speed of the workpiece W after the filtering process, and calculates a target value to be given to the servo driver 40 (step S116). Finally, the CPU unit 10 outputs a target value to be given to the servo driver 40 (step S118).

Then, when the next control cycle arrives, the processing after step S102 is executed again.
<F. Simulation results>
Next, an example of the result of verifying the effect related to the observer control shown in FIGS. 5 and 6 using simulation will be described.

  FIG. 7 is a diagram illustrating an example in which the improvement effect by the control system 100 according to the present embodiment is verified by simulation. The example shown in FIG. 7 simulates the behavior when an acceleration variation corresponding to 14% of the rated torque of the rotary cutter is given. (1) in FIG. 7 shows the actual speed (actual transport speed), and the more accurate the speed estimate is made as the estimated transport speed of the workpiece W becomes closer to the actual speed.

  (2) in FIG. 7 shows the speed calculated (difference) based on the detection signal (pulse signal) from the encoder 140, and (3) in FIG. 7 shows the filter element 14 when there is no observer control described above. The speed after smoothing output from is shown. As shown in FIG. 7 (3), it can be seen that a delay time is caused by the filter element 14.

  On the other hand, (4) of FIG. 7 shows the speed | velocity | rate after smoothing output from the filter element 14 when there exists the above-mentioned observer control. As shown in FIG. 7 (4), it can be seen that the delay time generated in the filter element 14 is compensated by the observer control.

  Thus, by applying the observer control, the delay time due to the filter element 14 can be reduced, and the effect caused by this can be minimized.

<G. First Modification>
In the above-described observer control, the configuration for compensating for the delay due to the filter element 14 has been described. On the other hand, the control system 100 as a whole tends to affect the output calculation due to the transmission delay of the pulse signal detected by the encoder 140, the dead time due to the control period in the CPU unit 10, and the like. The magnitude of such transmission delay and dead time due to the control cycle can be known in advance. Therefore, in order to compensate for such a predetermined dead time, the effect of delay can be avoided by introducing a phase shift filter in the above-described observer control. Hereinafter, a configuration with improved observer control will be exemplified.

  FIG. 8 is a schematic diagram illustrating a control structure mounted on the control device 1 according to the first modification. The control structure shown in FIG. 8 is realized by the CPU unit 10 executing a program. Basically, the CPU unit 10 executes arithmetic processing at a predetermined control cycle (for example, 10 msec). Therefore, the calculation related to the control structure shown in FIG. 8 is also repeatedly executed at a predetermined control cycle.

  The control structure shown in FIG. 8 differs from the control structure shown in FIG. 5 in that a phase shift filter 17 is added before the integration element 16. The phase shift filter 17 delays the phase by the designated shift amount with respect to the estimated conveyance speed of the workpiece W output from the filter element 14 and outputs the result to the integration element 16. The phase shift filter 17 compensates for a transmission delay, a dead time due to a control cycle, and the like. Basically, a phase to be delayed is determined based on a time comparable to these delay times. That is, the estimated conveyance speed of the workpiece W is delayed by a predetermined time and then fed back, so that the conveyance speed of the workpiece W is corrected in the direction in which the corresponding phase is advanced.

  The phase shift filter 17 is typically a first-order lag filter. The phase of the conveyance speed is shifted by a predetermined time constant (delay time) by the phase shift filter 17, and as a result, the phase of the processing reflection position is advanced by a system for calculating an observer correction value as described later. be able to.

  The time for shifting the phase in the phase shift filter 17 can be arbitrarily set according to the system. Typically, it is preferable that the time is substantially the same as the propagation delay time in the field network (and the time required for command value generation processing and output processing in the CPU unit 10).

  As described above, the observer control according to the first modification includes the phase shift filter 17 as means for shifting to the delay side with respect to the result of the smoothing process calculated in advance (the output of the filter element 14). The observer control according to the first modification is the difference between the measurement information (value read from the pulse counter unit 20: spindle position) and the result of the smoothing process with the phase shifted (output of the phase shift filter 17). As a means for calculating, a subtraction element 15 is included.

  According to the first modification, the future value can be estimated by shifting the phase of the delay observer and / or the dead time that can be predicted in the speed observer. As a result, it is possible to compensate for delays in the detection system and fluctuations in the target speed.

  As described above, the observer control according to the first modification includes the phase shift filter 17 as means for shifting to the delay side with respect to the result of the smoothing process calculated in advance (the output of the filter element 14). The observer control according to the first modification is the difference between the measurement information (value read from the pulse counter unit 20: spindle position) and the result of the smoothing process with the phase shifted (output of the phase shift filter 17). As a means for calculating, a subtraction element 15 is included.

  According to the first modification, the future value can be estimated by shifting the phase of the delay observer and / or the dead time that can be predicted in the speed observer. As a result, it is possible to compensate for delays in the detection system and fluctuations in target speed.

  Next, an example of the result of verifying the effect related to the observer control shown in FIG. 8 using simulation will be described.

  9 and 10 are diagrams illustrating an example in which the improvement effect by the control device 1 according to the first modification is verified by simulation. The example shown in FIGS. 9 and 10 is a simulation of the behavior when an acceleration variation corresponding to 14% of the rated torque of the rotary cutter is applied. However, it is assumed that there is a communication delay (2 ms) in the communication line from the encoder 140 to the control device 1.

  (1) in FIGS. 9 and 10 shows the actual speed (actual conveyance speed) of the encoder 140 input to the control device 1, and (1) ′ in FIGS. 9 and 10 indicates the actual speed (actual speed) of the encoder 140. Transport speed). That is, (1) in FIG. 9 and FIG. 10 corresponds to (1) ′ in FIG. 9 and FIG. 10 shifted by 2 ms, which is a dead time for communication. As the estimated conveyance speed of the workpiece W shows a value closer to (1) 'in FIGS. 9 and 10, the more accurate speed estimation is performed.

  (2) in FIGS. 9 and 10 show the speed calculated (difference) based on the detection signal (pulse signal) from the encoder 140, and (3) in FIGS. 9 and 10 shows the above-described observer control. The speed after smoothing output from the filter element 14 when there is not is shown.

  (4) of FIG. 9 and FIG. 10 shows the speed | velocity | rate after the smoothing output from the filter element 14 when there is observer control based on Embodiment 1. FIG. As shown in (4) of FIGS. 9 and 10, it can be seen that the delay time generated in the filter element 14 is compensated by the observer control. However, the dead time related to the communication of the detection signal of the encoder 140 cannot be compensated.

  On the other hand, (5) in FIG. 9 and FIG. 10 shows the speed after smoothing output from the filter element 14 when there is observer control further including a phase shift filter according to the first modification. Thus, by applying the observer control including the phase shift filter, it is possible to minimize the influence caused by the dead time related to communication in addition to the delay time due to the filter element 14.

<H. Second Modification>
In addition to the above-described control system 100, the following control system 200 can also be targeted as the spindle / slave-linked control application.

  FIG. 11 is a schematic diagram illustrating a configuration of a control system 200 according to the second modification. Referring to FIG. 11, control system 200 according to another embodiment of the present invention assumes a glass substrate of a solar panel or the like as a moving object (work W). The control system 200 forms a predetermined wiring pattern or the like by irradiating the workpiece W with laser light. The distance between the light source of the laser beam and the irradiation position of the workpiece W needs to be maintained at a predetermined value. Therefore, the distance to the workpiece W is measured by optical means, and the laser light source is sequentially moved in the vertical direction of the workpiece W based on the measured distance.

  More specifically, the laser light source 240 and the measurement camera 250 are disposed on the same substrate 220, and this substrate 220 is moved vertically by the servo motor 230, that is, perpendicular to the surface of the workpiece W. Moved to. By this movement, the distance between the substrate 220 and the laser light source 240 is maintained at a predetermined value. It is assumed that the surface of the workpiece W has irregularities that cannot be ignored.

  The surface of the workpiece W is imaged by the measurement camera 250, and the distance to the surface of the glass substrate is calculated by the glass surface distance calculation logic 260 from the image obtained by this imaging and the optical constant (focus position, etc.) at that time. The This distance is updated every predetermined period. That is, the distance measurement by the measurement camera 250 and the glass surface distance calculation logic 260 has a predictable dead time.

  The servo motor 230 is rotationally driven by being supplied with a signal (typically a pulse signal) related to the driving from the servo driver 40. In the example shown in FIG. 11, the measurement camera 250 and the glass surface distance calculation logic 260 are also referred to as a main axis, and the servo motor 230 that is driven by receiving information from the main axis is also referred to as a slave axis.

  The control device 1A includes a CPU (Central Processing Unit) unit 10 that executes main arithmetic processing, an AD (Analog to Digital) unit 50 that acquires distances output from the measurement camera 250 and the glass surface distance calculation logic 260, And a fieldbus interface unit 30 for communicating with field devices such as the servo driver 40.

  Since other basic configurations are the same as those of control system 100 shown in FIG. 1, detailed description will not be repeated.

  In the control system 200 shown in FIG. 11, since the information update cycle of the distance to the surface of the glass substrate is slower than the control cycle of the servo motor 230, and the distance information is defined by the analog signal, noise is included. There is a problem of getting. Therefore, if the output signal from the measurement camera 250 is directly used as a command value for the servo motor 230, a follow-up delay due to sampling delay and vibration due to sensing noise may occur. The former is dealt with by compensation of the phase advance time by introducing a phase shift filter, and the latter is dealt with by a low-pass filter (filter) process included in the observer control.

  FIG. 12 is a flowchart showing a processing procedure in the control system 200 according to the second modification. Each step shown in FIG. 12 is typically realized by the CPU unit 10 executing a program.

  Referring to FIG. 12, CPU unit 10 accepts a target distance setting value to the machining surface of workpiece W (step S <b> 200). Subsequently, the CPU unit 10 determines whether or not it is the image capturing timing of the measurement camera 250 (step S202). If it is not the timing of image capture by the measurement camera 250 (NO in step S202), the process of step S216 is executed.

  If it is the timing for capturing an image by the measurement camera 250 (YES in step S202), the CPU unit 10 captures the workpiece W by the measurement camera 250, and the workpiece W based on the captured image. Is obtained (step S204). Then, the CPU unit 10 calculates a deviation between the target distance and the acquired distance to the workpiece W (step S206).

  Subsequently, the CPU unit 10 calculates a time change amount with respect to the deviation from the target distance (step S208). Then, the CPU unit 10 corrects the time change amount for the deviation using the difference between the estimated deviation calculated in the previous control cycle and the deviation acquired in the current control cycle (step S210). Then, the CPU unit 10 performs a filtering process on the time variation with respect to the corrected deviation (step S212).

  Subsequently, the CPU unit 10 calculates an estimated deviation from the amount of time change using the deviation after filtering, and calculates a target value to be given to the servo driver 40 (step S214). Specifically, the target position to be given to the servo driver 40 is determined by adding an estimated deviation to the target distance setting value. Finally, the CPU unit 10 outputs a target value to be given to the servo driver 40 (step S216).

Then, when the next control cycle arrives, the processing after step S202 is executed again.
Here, the process of estimating the future value using the observer control, that is, predicting the future target position will be described in more detail.

  FIG. 13 is a schematic diagram illustrating the entire control structure implemented in the control system according to the second modification. Referring to FIG. 13, control device 1A includes a future position prediction unit 270 as its control structure. The control structure inside the future position prediction unit 270 is substantially the same as the control structure shown in FIG. The future position prediction unit 270 receives the input data and the phase shift amount, and outputs a future value of the target position information. The future position prediction unit 270 receives the board deviation e (k, 0). This substrate deviation e (k, 0) is input as the spindle position in FIG. Here, the first variable (k) in parentheses indicates control timing (control cycle number), and the second variable in parentheses indicates control timing (phase).

  The first variable in parentheses included in the phase shift amount S (1, m) indicates an imaging interval by the measurement camera 250, and the second variable in parentheses is a delay time due to transmission on the network and a CPU. The total time (dead time) of the processing time in the unit 10 is shown.

The future position predicting unit 270 outputs the predicted substrate deviation predicted value e (k + 1, m) ^* . Here, the first variable in parentheses indicates control timing (control cycle number), and the second variable in parentheses is the total time of delay time due to transmission on the network and processing time in the CPU unit 10. (Dead time).

Hereinafter, the effects according to the second modification will be described in comparison with related technologies.
FIG. 14 is a diagram illustrating an example of the effect of target position prediction related to the control system according to the second modification. FIG. 15 is a diagram illustrating an example of another target position prediction effect related to the control system according to the second modification. FIG. 16 is a diagram illustrating an example of the effect of target position prediction by the control system according to another form of the second modified example.

  First, referring to FIG. 14, the deviation information acquisition timing (distance information acquisition timing by the measurement camera 250) and the servo control timing (timing at which the command value is actually given to the servo driver 40) are synchronized. Is not limited. Therefore, if the target position is set without considering any information on t (k, 0), this setting is reflected at t (k, m). That is, the target position is switched at the timing shown by the thick line in FIG. As a result, the target position becomes constant during the deviation information acquisition timing, and vibration (displacement) increases at the target position switching timing.

  FIG. 15 shows a technique capable of mitigating the vibration generated at the target position in the control configuration shown in FIG. More specifically, for example, by using an interpolation logic such as backward difference approximation, the temporal change (trajectory) of the target position becomes smooth. However, the method of FIG. 15 has a problem that the deviation (following error) between the substrate deviation and the target position e is increased.

On the other hand, in this embodiment, as shown in FIG. 16, first, a phase deviation S (1, m) is given, so that a future deviation is predicted. More specifically, the future deviation prediction unit 170 receives the phase shift amount S (1, m) corresponding to the prediction time, and outputs a substrate deviation predicted value e (k + 1, m) ^* . Further, the interpolation unit 172 executes interpolation logic such as backward difference approximation for each control cycle based on the substrate deviation predicted value e (k + 1, m) ^* . Thereby, noise included in the substrate deviation e can be removed. In addition, by predicting future values, it is possible to generate a target trajectory that is smooth in time.

  As described above, the future position predicting unit 270 predicts the measurement information based on the acquired temporal change of the measurement information.

<I. Third Modification>
In the above-described embodiment, the observer control has been described as an example of compensation means for compensating for a time delay between the physical displacement of the object (main axis: rotational speed and displacement) and the acquired measurement information. . On the other hand, in the third modification, a method for compensating for a delay that occurs in acquired measurement information by predicting a physical displacement of an object based on past results will be described. More specifically, delay information (hint data) that can be predicted in advance is generated, and a position (future position) ahead by a predetermined time from a certain point in time is determined using the generated hint data.

  As a third modification, a control structure mounted on the control device 1A of the control system 200 shown in FIG. 11 will be described.

  FIG. 17 is a schematic diagram illustrating the entire control structure implemented in the control system according to the third modification. The control structure shown in FIG. 17 is realized by the CPU unit 10 executing a program. Basically, the CPU unit 10 executes arithmetic processing at a predetermined control cycle (for example, 10 msec). Therefore, the calculation according to the control structure shown in FIG. 17 is also repeatedly executed at a predetermined control cycle.

  Referring to FIG. 17, control device 1A includes a future position prediction unit 150 as its control structure. The control structure inside the future position prediction unit 150 is substantially the same as the control structure shown in FIG. The future position prediction unit 150 accepts, as input data, a value obtained by adding hint data to the current measurement result (measurement result acquired in the current control cycle), and sets the phase shift time set in the phase shift filter 17 as a set value Accept as. The input data is input as the spindle position in FIG. Then, the future position prediction unit 150 outputs the input data and the future position.

  As the phase shift time, a predictable value such as a detection delay of the detector or a delay time on the network is set.

  FIG. 18 is a diagram for explaining a method of creating hint data included in input data input to the control device 1A according to the third modification. FIG. 19 is a diagram illustrating an example of the effect of future position prediction by the control device 1A according to the third modification. FIG. 19 shows a result of predicting the position of the workpiece W after 16 msec in the control system 200 shown in FIG. 11 when the sampling cycle of the measurement camera 250 is 2 msec.

  Referring to FIG. 19, the hint data is data obtained by shifting the previous measurement result (measurement result acquired in the previous control cycle) by the dead time as the correct data, and the correct data is used as the previous data. The subtraction of the measurement result is used as hint data. Here, the previous measurement result is measured with a certain delay time. By using the difference between the two as hint data, the tracking error can be reduced. That is, the future position prediction unit 150 as a compensation unit generates hint data from the difference between the measurement information obtained by shifting the measurement information in time by the time delay generated in the control system and the current measurement information.

  FIG. 19 shows the correct answer value (solid line) and the current measurement result measured by the measurement camera 250. Furthermore, hint data and the result of filtering hint data are shown. A position closer to the correct value can be predicted by correcting the measured value using the hint data or the filtered hint data (this method). FIG. 19 also shows the positions predicted by the method related to this embodiment. In particular, it can be seen that the position can be predicted more accurately according to the present embodiment in the range of large fluctuations.

  FIG. 19 shows the error of the predicted value with respect to the correct value. FIG. 19 shows a comparison between the case where the prediction method using hint data according to the third modification is used and the case where it is not. As shown in FIG. 19, position prediction with less error can be performed by compensating the delay using the hint data according to the present embodiment.

Thus, in the third modified example, the command value is generated using the hint data.
<J. Fourth Modification>
In the above-described embodiment, the method of compensating for the delay that occurs in the acquired measurement information by predicting the physical displacement of the object based on the past performance has been described. On the other hand, in the third embodiment, a method for determining future data by extrapolation using information acquired for the same workpiece will be described.

  As a fourth modification, a control structure mounted on the control device 1 of the control system 100 shown in FIG. 1 will be described.

  FIG. 20 is a schematic diagram illustrating a control structure mounted on the control device 1 according to the fourth modification. FIG. 21 is a diagram for explaining a principle for performing time delay compensation in the control structure shown in FIG.

  Referring to FIG. 20, control device 1 includes a buffer 160, an interpolation unit 162, a prediction unit 164, and a control unit 166 as its control structure.

  The buffer 160 temporarily stores the spindle position measured by the measuring roll (encoder 140). The interpolation unit 162 refers to the spindle position and the like stored in the buffer 160 and interpolates (extrapolates) these to calculate the previous position for a predetermined period. Typically, linear interpolation (primary interpolation), various multidimensional interpolations, and the like can be used.

  The prediction unit 164 calculates the predicted position by combining the measured spindle position and the previous position for a predetermined period calculated by the interpolation unit 162. The calculated predicted position is input to the control unit 166. The control unit 166 performs control based on the input predicted position, and calculates a command value given to the servo motor 110. As the control unit 166, general PID control may be employed.

  Next, a method for calculating a predicted position will be described with reference to FIG. Like the arrow shown in FIG. 21, the change of the future is estimated only for a predetermined period using several points of the position acquired previously. Thus, if the change of the position over a certain period can be acquired, the previous change can be predicted. Since the physical displacement of the object can be predicted in this way, even if there is a time delay in the acquired information, an appropriate command value can be generated and output without being affected by the time delay.

  Thus, in the fourth modified example, the prediction unit 164 predicts the measurement information based on the temporal change in which the measurement information is acquired.

<K. Advantage>
According to the control device according to the present embodiment, it is possible to reduce the influence of quantization error, noise, and the like included in the physical displacement of the workpiece W by using filter processing (smoothing processing). At the same time, the influence of the delay time caused by the filter processing is corrected using observer control. As a result, highly accurate control can be realized while suppressing the influence of quantization error and noise.

  More specifically, by applying the observer control to the command value for the servo motor 110, the time of the command value can be advanced by the sampling delay, so that the responsiveness can be improved without impairing the stability. it can.

  Furthermore, by introducing a phase shift filter to the observer control, it is possible to reduce the influence of transmission delay and measurement system dead time. This eliminates the effect of a predictable dead time.

  More specifically, since the phase advance time in the phase shift filter can be explicitly specified, phase adjustment in feedforward control and / or feedback control is easily performed in a system in which dead time such as a field network can be clearly determined. be able to.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  DESCRIPTION OF SYMBOLS 1 Control apparatus, 2 spindle apparatus, 3 slave axis apparatus, 10 CPU unit, 11 differentiation element, 12 amplification element, 13 addition element, 14 filter element, 15 subtraction element, 16, 18 integration element, 17 phase shift filter, 19 target Value calculation unit, 19a cam table, 20 pulse counter unit, 30 fieldbus interface unit, 40 servo driver, 100, 200 control system, 102, 104 rotary cutter, 110, 230 servo motor, 112, 114 transport motor, 140 encoder, 220 substrate, 240 laser light source, 250 measurement camera, 260 glass surface distance calculation logic, W work.

Claims (7)

An input unit for acquiring measurement information indicating physical displacement of the object from the measurement device;
An interface that communicates with a drive unit for driving a processing device that performs processing on the object;
A control unit for controlling the drive unit based on the measurement information,
The controller is
Smoothing means for smoothing the measurement information;
A correction unit that corrects the measurement information before the smoothing process by comparing a result of the smoothing process calculated in advance with the measurement information, and the correction unit includes:
Means for shifting the phase to the delay side with respect to the result of the smoothing process calculated in advance;
A control device including means for calculating a difference between the measurement information and a result of the smoothing process in which the phase is shifted. The measuring device measures the position at each time point related to the movement of the object as the physical displacement,
The control device according to claim 1 , wherein the smoothing unit smoothes a moving speed that is a temporal change amount of the position of the object. At least one input unit and the drive unit, exchanging information through the network control device according to claim 1 or 2. The said control part is a control apparatus of any one of Claims 1-3 which performs a calculation process with a predetermined period. The control unit further includes calculation means for calculating a command value to be given to the drive unit based on the result of the smoothing process,
Said calculation means, said estimates the position of the object from the results of the smoothing process, the control device according to any one of claims 1-4. An input unit that acquires measurement information indicating physical displacement of an object from a measurement device, an interface that communicates with a drive unit for driving a processing device that performs processing on the object, and a control unit A control program executed on a computer, the control program being
Performing a smoothing process on the measurement information;
By comparing the measurement information with the result of the smoothing process calculated in advance, the step of correcting the measurement information before the smoothing process,
The correcting step includes
Shifting the phase to the lag side with respect to the result of the smoothing process calculated in advance;
Calculating a difference between the measurement information and a result of the smoothing process in which the phase is shifted. An input unit that acquires measurement information indicating physical displacement of an object from a measurement device, an interface that communicates with a drive unit for driving a processing device that performs processing on the object, and a control unit A control method executed by a computer,
Performing a smoothing process on the measurement information;
A step of correcting the measurement information before the smoothing process by comparing the measurement information with the result of the smoothing process calculated in advance.
The correcting step includes
Shifting the phase to the lag side with respect to the result of the smoothing process calculated in advance;
Calculating a difference between the measurement information and a result of the smoothing process in which the phase is shifted.

Images