JPWO2015132872A1 - Feedback control method, feedback control device, and program - Google Patents

Abstract

The object of the present invention is to shorten the time until a control target converges to a target value, to make it difficult for the control target to deviate from the target value, and to prevent deterioration of the stability of the control system. A first calculation result including a low-speed integration calculation result is obtained based on a deviation between a target signal and a detection signal to be controlled, and a power unit is driven by the first calculation result. In the feedback control method of controlling the control target and using the control result of the control target as the detection signal, a fast integration calculation of the deviation is performed when the absolute value of the deviation is within a preset first threshold value. Then, the integrated value is added to the first calculation result as the second calculation result, or the fast integration calculation is performed on the first calculation result to set the integration value as the second calculation result, The power unit is driven by the second calculation result. [Selection] Figure 1

Description

  The present invention relates to a feedback control method, a feedback control device, and a program for controlling a web tension and other controlled objects in a predetermined state.

  FIG. 5 shows an example of a web tension control device. This is a device that applies a predetermined tension to the web 10 during conveyance in order to prevent the web 10 such as printing paper to be rewound from the roll state from being slackened or wrinkled. FIG. 6 shows a functional block diagram of a feedback control device used in the tension control device.

  The web 10 is rewound in the direction of arrow A from the roll state with the two rollers 20 and the intermediate roller 30 interposed therebetween. Tension is applied to the web 10 by downstream winding power and upstream rewind braking force. The force applied to the intermediate roller 30 by the web 10 is a force corresponding to this tension, and is detected by the two tension detectors 40 arranged to support both ends of the intermediate roller 30 and taken into the controller 50B. The controller 50B has a target tension value set in advance. The target tension signal and a tension detection signal obtained by converting the tension detection signal (force signal) detected by the tension detector 40 into an actual tension. A control signal corresponding to the deviation is generated by the controller 50B. By this control signal, the electromagnetic brake 60 as the power unit is driven, so that the braking force for rewinding the roll of the web 10 is controlled, and the tension applied to the web 10 is controlled to be the target tension. The In addition to the electromagnetic brake 60, a clutch or a motor is also used as the power unit. In recent years, there are many cases where motor control is used, but in terms of cost, the use of the electromagnetic brake tends to be cheaper. .

  FIG. 7 shows the internal configuration of the controller 50B. The substance of the tension detection signal (force signal) detected by the tension detector 40 is the force applied to the intermediate roller 30, and this signal is converted to the original value in the signal converter 53 based on the calibration value given in advance. Converted to a tension signal. When the noise is large, the switch SW1 is switched to the broken line side in advance to remove the noise by the low-pass filter 51 and then input to the signal converter 53. The target tension signal is subtracted from the tension detection signal output from the signal converter 53 in the subtractor 54 to become a deviation signal. The deviation signal is taken into the PID calculator 55 to perform proportional / integral / differential calculations.

  Since the basic principle of the tension detector 40 is to detect the pressure applied to the intermediate roller by taking the tension of the web 10 as a spring displacement, the relationship between the mass of the intermediate roller 30 and the spring constant of the tension detector 40 is used. May have a resonance point of several tens of Hz to several tens of Hz. For this reason, it is desirable to cut the signal component at the resonance point with certainty. Therefore, a first-order lag low-pass filter having a time constant of 10 milliseconds to several hundred milliseconds is inserted as the low-pass filter 51. In recent years, the tension detector 40 using a strain gauge has appeared. Since this type of tension detector has a very large spring constant, the resonance frequency of the tension detector 40 is increased, so that the time constant of the low-pass filter 51 can be reduced.

  In the PID calculation unit 55, a proportional calculator 551 that performs a proportional calculation has a role of proportionally amplifying the deviation and quickly correcting the deviation. If only the proportional calculation is performed, if the deviation becomes small, the result of the control calculation becomes close to zero, so that a steady deviation remains and a sufficient control calculation cannot be realized. Therefore, when the integral calculation of the deviation up to the present is performed by the integral calculator 552, the drawback is improved. In the integral calculation, even if the deviation is small, it is integrated, so that it is possible to produce an output that eliminates the steady deviation of the controlled object. In general, the gain of integration is called integration time. The shorter the integration time, the higher the gain. In many cases, sufficient control characteristics can be obtained only by proportional and integral calculations, and this control based only on proportional and integral calculations is called PI control. The differential calculation performed by the differential calculator 553 works effectively against sudden deviation fluctuations. The calculation results of the calculators 551, 552, and 553 are added by the adder 554 to be a control output signal. When the noise is large, the deviation signal is input to the PID calculator 55 after passing through the low-pass filter 52 by switching the switch SW2 to the broken line side in advance. The calculation in the controller 50B is performed by adjusting each parameter of the PID.

  In feedback control for causing a control target to follow a target value, it is common to insert at least one integral calculator in the controller 50B. In design theory, the calculation result derived by the controller 50B may not include an integration result. In this case, it is assumed that the control object includes an integration operation, and the integration calculation is transferred to the controller 50B in the implementation. It is common. This is because it is theoretically known that the steady deviation is eliminated by inserting one or more integration calculators into the controller 50B (internal model principle).

  By the way, generally, increasing the gain of the control calculation of the controller 50B has many advantages such as a faster response of the control system and a decrease in steady-state deviation, but this lowers the stability of the control system. Therefore, the gain cannot be increased beyond a certain value.

  For example, in a rotary printing press, the mass of the unwinding roll continues to change. In the tension control of this rotary printing press, if the roll mass changes with time, the optimum control parameter also changes with time. In this case, it is necessary to adjust the control parameter to be stable under all conceivable conditions, but generally, stability is ensured by setting the gain of the controller 50B low. For this reason, it is difficult to construct a highly accurate control system.

  Several proposals have been made and implemented for this issue. Since the control output of the controller 50B tends to be proportional to the roll mass, the output when the roll mass is small (the brake control output signal is small) is smaller and the output when the roll mass is large (the brake control output signal is large). There is an example of exponent correction to make larger. This method can be implemented with only an analog circuit by utilizing the nonlinear characteristics of the diode. At present, microprocessors, FPGAs, and the like are available at low cost, and there are many examples of adopting them for tension control. For this reason, in many cases, this function can be implemented at no additional cost.

  However, this method does not completely cancel the roll weight change, and cannot guarantee whether the gain of the control calculation of the controller 50B can be increased to the optimum value. Therefore, even if this exponent correction is used, the gain of the control calculation of the controller 50B cannot be sufficiently increased, and there is a possibility that disturbance suppression and control accuracy cannot be sufficiently increased.

  As a more advanced means, as shown by a controller 50C in FIG. 8, the calculation by the diameter correction unit 59 can be added to the calculation result of the PID calculator 55 by the switch SW4. An exponent correction factor is applied to the signal. This is obtained by calculating or measuring the winding diameter of the rewinding roll, and reflecting this in the control parameter of the diameter correction unit 59. The unwinding roll diameter can be easily obtained by determining the surface position of the unwinding roll with a contact or non-contact position detector. Further, the diameter of the rewinding roll can be easily calculated by using an encoder for obtaining the line speed and an encoder (or a proximity sensor) for measuring the rotation of the rewinding roll.

  However, it can be easily imagined that the method using the diameter correction unit 59 has two drawbacks. One is an increase in the price of the system. The other is that the square of the roll diameter and the moment of inertia of the roll are in a proportional relationship, but if the roll diameter is known, the moment of inertia of the roll is not known. For example, if the paper width or the paper weight density changes, the mass varies even with the same roll diameter. Also, if the roll winding tension is different, the mass and moment of inertia may change even with the same roll diameter. That is, simply applying the roll diameter calculation result and setting the parameters of the diameter correction unit 59 to realize the variable gain controller 50C is not guaranteed to be the optimum control parameter, and there is some stability margin. Must be held.

  Naturally, it can be estimated that the problem can be solved by realizing a system in which the moment of inertia of the roll is estimated in real time and the control parameter of the diameter correction unit 59 is varied according to each state. However, in order to input such information every time the rotary printing machine performs printing, it is necessary to construct a huge database system, which is not preferable for operation. Moreover, it takes a lot of time and effort to make such a system on the machine manufacturer side that supplies the apparatus.

  Another possible means is to use a robust control theory (for example, Patent Document 1). This is a method for obtaining a mathematical model of a controlled object, determining a variation range of the controlled object, and obtaining an optimal control calculation.

JP 2010-51104 A

  However, this method requires advanced control theory to adjust the control parameters. For this reason, it cannot be said that it is realistic to employ this for a controller that is intended for a normal factory worker to adjust the control parameters. There is also a problem in adopting robust control as a dedicated controller. This is because robust control design / adjustment requires knowledge of advanced control theory and expensive design software. For this reason, it is thought that the control object which can employ | adopt robust control theory is limited to the thing which requires the effort and cost to control itself.

  In order to commercialize a general-purpose controller, it must be assumed that ordinary factory workers will be able to use it. They do not necessarily have advanced control theory. Even if there is an ability to make full use of advanced control theory, it is not always permitted in cost to introduce and operate expensive design software.

  Further, it is theoretically known that when the target value changes in a ramp shape, the system cannot be terminated at the target value in a system having only one integrator (internal model principle). This problem can be solved by using a plurality of integrators, but the integrators cannot induce a high controller gain because they induce a phase delay.

  In general, instead of using a single integrator in the controller, increase the integral gain to reduce the steady-state deviation value, or set the integral gain high to always overshoot, etc. However, depending on the control target, as described above, the gain of the control calculation of the controller may not be sufficiently increased. In this case, the required performance may not be met.

  An object of the present invention is to reduce the time until the control target converges to the target value, to make it difficult for the control target to deviate from the target value, and to prevent deterioration of the stability of the control system. It is to provide a feedback control device and a program.

  In order to achieve the above object, a feedback control method according to a first aspect of the present invention obtains a first calculation result including a low-speed integration calculation result based on a deviation between a target signal and a detection signal to be controlled, In the feedback control method in which the power unit is driven by the first calculation result, the control target is controlled by the power unit, and the control result of the control target is the detection signal, the absolute value of the deviation is set in advance. When the difference is within a threshold value of 1, the deviation is subjected to a high-speed integration operation, and the integration value is added to the first operation result to obtain a second operation result, or the high-speed operation is performed with respect to the first operation result. An integral calculation is performed, and the integral value is set as a second calculation result, and the power unit is driven by the second calculation result.

  According to a second aspect of the present invention, in the feedback control method according to the first aspect, when the absolute value of the deviation exceeds the first threshold value, the integral value of the high-speed integral calculation is changed to the integral value of the low-speed integral calculation. The high-speed integral calculation is reset to zero, and the high-speed integral calculation is stopped.

  The invention according to claim 3 is the feedback control method according to claim 1 or 2, wherein when the deviation can be regarded as zero or zero, the integral value of the fast integration operation is added to the integration value of the slow integration operation, In addition, the high-speed integration calculation is continued after the integration value of the high-speed integration calculation is reset to zero.

  According to a fourth aspect of the present invention, in the feedback control method according to the first or second aspect, a second threshold value smaller than the first threshold value is provided, and when the absolute value of the deviation is within the second threshold value, The integral value of the high-speed integration calculation is added to or not added to the integration value of the low-speed integration calculation, and the high-speed integration calculation is stopped after the integration value of the high-speed integration calculation is reset to zero.

  According to a fifth aspect of the present invention, in the feedback control method according to the first, second, third, or fourth aspect, the maximum integration value of the high-speed integration calculation is set smaller than the maximum integration value of the low-speed integration calculation. And

  According to a sixth aspect of the present invention, there is provided a feedback control device including a first calculation unit including a low-speed integration calculator that performs low-speed integration based on a deviation between a target signal and a detection signal to be controlled, and the first calculation unit. A power unit that is driven by the obtained first calculation result to control the control object; and a detector that detects the control result of the control object, and the signal detected by the detector is the detection signal. In the feedback control apparatus, when the absolute value of the deviation is within a preset first threshold value, a high-speed integration calculation of the deviation is performed, and the integration value is added to the first calculation result to obtain a second value. A high-speed integration computing unit is provided as a calculation result, or a high-speed integration calculation is performed on the first calculation result and the integration value is used as a new second calculation result. To drive And features.

  According to a seventh aspect of the present invention, in the feedback control device according to the sixth aspect, when the absolute value of the deviation exceeds the first threshold value, the integral value of the fast integral computing unit is changed to that of the slow integral computing unit. The integral value of the high-speed integration calculator is reset to zero and the operation of the high-speed integration calculator is stopped.

  According to an eighth aspect of the present invention, in the feedback control device according to the sixth or seventh aspect, when the deviation can be regarded as zero or zero, the integral value of the fast integral computing unit is added to the integral value of the slow integral computing unit. In addition, the high-speed integration arithmetic unit is continuously operated after the integral value of the high-speed integration arithmetic unit is reset to zero.

  The invention according to claim 9 is the feedback control device according to claim 6 or 7, wherein a second threshold value smaller than the first threshold value is provided, and when the absolute value of the deviation is within the second threshold value, The integral value of the high-speed integral computing unit is added to the integral value of the low-speed integral computing unit, and the integral value of the high-speed integral computing unit is reset to zero, and then the high-speed integral computing unit is stopped. .

  According to a tenth aspect of the present invention, in the feedback control device according to the sixth, seventh, eighth, or ninth aspect, the maximum integration value of the high-speed integration calculator is set smaller than the maximum integration value of the low-speed integration calculator. It is characterized by that.

  A program according to an eleventh aspect of the invention acquires a first calculation result including a low-speed integration calculation result based on a deviation between a target signal and a detection signal to be controlled, and drives a power unit by the first calculation result. In the program for controlling the control object by the power unit and causing the computer to perform an integration operation in a feedback control method using the control result of the control object as the detection signal, the absolute value of the deviation is set in advance. A first step for determining whether or not the difference is within a threshold value, and when it is determined by the first step that the absolute value of the deviation is within a first threshold value set in advance, the fast integration of the deviation An operation is performed and the integration value is added to the first operation result to obtain a second operation result, or the fast integration operation is performed on the first operation result and the integration value is converted to the second operation result. Comprising a second step of the operation result, and the power unit by the second operation result is characterized by being so driven.

  The invention according to claim 12 is the program according to claim 11, wherein when the absolute value of the deviation exceeds the first threshold value, the integration value of the fast integration operation is added to the integration value of the slow integration operation. Or a third step of resetting the integration value of the high-speed integration calculation to zero and stopping the high-speed integration calculation.

  The invention according to claim 13 is the program according to claim 11 or 12, wherein when the deviation can be regarded as zero or zero, the integral value of the fast integral operation is added to the integral value of the slow integral operation, and A fourth step of continuing the high-speed integration calculation after resetting the integration value of the high-speed integration calculation to zero is provided.

  The invention according to claim 14 is the program according to claim 11 or 12, wherein a second threshold value smaller than the first threshold value is provided, and the fast integration is performed when the absolute value of the deviation is within the second threshold value. A fifth step of adding the integration value of the calculation to the integration value of the low-speed integration calculation and resetting the integration value of the high-speed integration calculation to zero and then stopping the high-speed integration calculation is provided.

  According to a fifteenth aspect of the invention, in the program according to the eleventh, twelfth, thirteenth or fourteenth aspect, the maximum integration value of the high-speed integration operation is set smaller than the maximum integration value of the low-speed integration operation. .

  In Claims 1, 6, and 11, since the high-speed integration calculation is performed only when the absolute value of the deviation is within the threshold value, the time until the control target converges to the target value while maintaining the stability of the control system. The control object can be made difficult to deviate from the target value.

  According to the second, seventh, and twelfth aspects, when the absolute value of the deviation exceeds the threshold value, the high-speed integration value is reset to zero and the high-speed integration calculation is stopped. Therefore, the next high-speed integration starts from zero. At this time, if the previous high-speed integral value is added to the low-speed integral value, a sudden change in the manipulated variable is prevented.

According to the third, eighth, and thirteenth aspects, when the deviation becomes zero, the high-speed integration value is reset to zero, so that the effect of the high-speed integration is reduced, and overshoot is less likely to occur. At this time, if the previous high-speed integral value is added to the low-speed integral value, a sudden change in the manipulated variable is prevented. As a result, the same effect can be obtained as when the low-speed integration is temporarily increased in speed.
Claims 4, 8 and 14 are essentially the same as claim 3. This means that a dead zone is set in the control system.

  According to the fifth, tenth and fifteenth aspects, the maximum integration value of the high-speed integration is set smaller than the maximum value of the low-speed integration, so that the operation amount by the high-speed integration can be suppressed to be small and the control stability can be ensured.

It is a functional block diagram which shows the structure of the controller of the feedback control apparatus of 1st Example of this invention. It is a flowchart of the high-speed integration calculation of the controller of a 1st Example. It is a functional block diagram which shows the structure of the controller of the feedback control apparatus of the 2nd Example of this invention. It is a flowchart of the high-speed integration calculation of the controller of the 3rd Example of this invention. It is a figure which shows the structure of the tension control apparatus which controls the tension of a web. It is a functional block diagram of the feedback control apparatus used with the tension control apparatus of FIG. It is a functional block diagram which shows the structure of the controller of the feedback control apparatus of FIG. FIG. 6 is a functional block diagram of another example controller used in the tension control device of FIG. 5.

  It is desirable to solve the problem of controllability in feedback control by using a control calculation that can be understood by a normal factory worker as an extension of the PID calculation. In the case of a control system in which the output of the power unit maintains a constant value when the deviation converges to zero, such as web tension control, it is considered that the output of the controller and the integrated value coincide. In this case, the time required for convergence can be shortened by increasing the response of the integrator. However, if the integration time is decreased in order to increase the response, the stability of the control system deteriorates.

  Therefore, in the present invention, in the feedback control applied to the web tension control device, the integral computing unit included in the conventional PID computing unit is treated as a low-speed integrating computing unit with slow response (large integration time). Introducing a new high-speed integration calculator with fast response. With respect to the high-speed integration arithmetic unit, the stability of the control system is not deteriorated by limiting the timing of the arithmetic operation.

<First embodiment>
FIG. 1 shows a controller 50 for feedback control according to a first embodiment of the present invention. The controller 50 is incorporated in the feedback control device described with reference to FIGS. 5 and 6 in place of the controllers 50B and 50C.

  The controller 50 of this embodiment includes low-pass filters 51 and 52, a signal converter 53 that converts a tension detection signal indicating force into an actual tension signal, a subtractor 54, a PID calculation unit 55, and a high-speed integration calculator with a short integration time. 56, an adder 57, and switches SW1, SW2, and SW3. The PID calculation unit 55 includes a proportional calculator 551, an integral calculator 552, a differential calculator 553, and an adder 554 that adds the calculation results of the calculators 551 to 553. Here, as the integration calculator 552 in the PID calculation unit 55, a low-speed integration calculator having a long integration time is used.

  The tension detection signal detected by the tension detector 40 is input to the signal converter 53 via the switch SW1, but when the noise is large, the switch SW1 is switched to the broken line side in advance, so that the low-pass filter 51 After the noise is suppressed, the signal is input to the signal converter 53. The tension detection signal (force signal) input to the signal converter 53 is converted from a force signal to a tension signal and output. The subtractor 54 generates a deviation signal having polarity by subtracting the tension detection signal from the input target tension signal. The deviation signal is input to the PID calculation unit 55 via the switch SW2. When the noise is large, the deviation signal is input to the PID calculator 55 after the noise is suppressed by the low-pass filter 52 by switching the switch SW2 to the broken line side in advance. When the switch SW3 is on, the signal is also input to the fast integration calculator 56. The calculation results in the PID calculator 55 and the fast integration calculator 56 are added by the adder 57 and output as a control output signal.

  The switch SW3 is turned on when the absolute value of the deviation signal obtained by the subtracter 54 is within a predetermined first threshold value, and is turned off when the absolute value exceeds the first threshold value. When this is off, the high-speed integration calculator 56 stops the integration calculation and resets the previous integration value (hereinafter referred to as “high-speed integration value”) to zero. Further, when the deviation signal becomes zero (crosses the zero point) or becomes small enough to be regarded as zero, the fast integration calculator 56 resets the fast integration value up to that point to zero. When resetting the high-speed integral value to zero, the high-speed integral value may be added to the integral value of the low-speed integral computing unit 552 (hereinafter referred to as “low-speed integral value”) before the zero reset. Also good.

  FIG. 2 shows a flowchart of a control program for the high-speed integration calculator 56. In order to realize the processing of the controller 50 according to the present embodiment, a digital arithmetic device using a computer such as a microprocessor or FPGA is used, and processing is performed every sampling of the deviation obtained by the subtractor 54. When high-speed integration is performed by the high-speed integration calculator 552, the switch SW3 is turned on.

  First, it is determined whether or not the absolute value of the sampled deviation is within the first threshold (step S1). When the absolute value of the deviation is within the first threshold, it is determined whether or not the deviation is zero or zero (step S2). At this time, if the deviation can be regarded as zero or zero, after adding the high-speed integral value of the deviation obtained in the previous sampling to the low-speed integral value, reset the high-speed integral value to zero, The current deviation (deviation that can be regarded as zero or zero) is rapidly integrated (step S3), and the high-speed integrated value is added to the PID calculation result (step S4). If the deviation cannot be regarded as zero or zero, the deviation is rapidly integrated (step S5), and the obtained high-speed integrated value is added to the PID calculation result (step S4). On the other hand, when the absolute value of the deviation is outside the first threshold range in step S1, the high-speed integral value is calculated after adding the high-speed integral value of the deviation obtained in the previous sampling to the low-speed integral value. After resetting to zero, stop high-speed integration. At this time, the fast integration value added to the PID calculation result in step S4 is zero. Next, when the sampled deviation is taken in, the same processing as described above is repeated according to the absolute value of the deviation. As described above, when the high-speed integration value is reset to zero, the high-speed integration value is not necessarily added to the low-speed integration value.

  By performing such control, high-speed integration is performed only when the absolute value of the deviation is within the first threshold, so that the control target is set to the target value while preventing the control system from running out of control. The time until convergence is shortened, and the controlled object is less likely to deviate from the target value. When the deviation becomes zero or a value that can be regarded as zero, the high-speed integration value is reset to zero, so that the effect of the high-speed integration is reduced, and overshoot is less likely to occur. Further, when resetting the high-speed integral value to zero, the rapid change of the operation amount is prevented by adding the high-speed integral value to the low-speed integral value immediately before that. This is the same as temporarily increasing the speed of the low-speed integration calculator. Note that, by setting the maximum integration value of the high-speed integration calculator smaller than the maximum integration value of the low-speed integration calculator, the amount of operation by the high-speed integration calculator is reduced.

<Second embodiment>
FIG. 3 shows a functional block diagram of the controller 50A of the feedback control apparatus of the second embodiment. A difference from the controller 50 described in FIG. 1 is that a high-speed integration calculator 56 is provided on the output side of the PID calculator 55 via the switch SW3, and the calculation result of the high-speed integration calculator 56 is PID calculated by the adder 58. It is the point which comprised so that the calculation result by the device 55 might be added.

  Also in this embodiment, in the same manner as the controller 50 described in FIG. 1, the high-speed integration is performed only when the absolute value of the deviation is within the first threshold range by controlling with the control program described in FIG. , Added to the PID calculation result. For this reason, there is an effect similar to that of the feedback control of the first embodiment such that the time until the control target converges to the target value can be shortened while preventing the control system from running away.

<Third embodiment>
FIG. 4 shows a flowchart of a control program of the high-speed integration computing unit 56 of the feedback control apparatus of the third embodiment. This embodiment can be applied to the controller 50 of FIG. 1 and the controller 50A of FIG. In the present embodiment, a first threshold value and a second threshold value smaller than the first threshold value are set, and only when the absolute value of the deviation is between the first and second threshold values, the fast integration computing unit 56 Integral operation was performed. When fast integration is performed, the switch SW3 is turned on.

  First, it is determined whether or not the absolute value of the sampled deviation is within a first threshold (step S11). If the absolute value of the deviation is within the first threshold value, it is next determined whether or not the absolute value of the deviation is within the second threshold value (step S12). If the absolute value of the deviation is within the second threshold, after adding the high-speed integral value of the deviation obtained in the previous sampling to the low-speed integral value, reset the high-speed integral value to zero, and then perform the high-speed integration. Stop (step S13). At this time, the fast integration value added to the PID calculation result in step S14 is zero. If the absolute value of the deviation is not within the second threshold, the deviation is rapidly integrated (step S15), and the obtained high-speed integrated value is added to the PID calculation result (step S14). On the other hand, if the absolute value of the deviation is outside the first threshold range in step S11, the high-speed integral value obtained by the previous sampling is added to the low-speed integral value, and then the high-speed integral value is set to zero. Fast integration stops after reset. At this time, the fast integration value added to the PID calculation result in step S14 is zero. Next, when the sampled deviation is taken in, the same processing as described above is repeated according to the absolute value of the deviation. As described above, when the high-speed integration value is reset to zero, the high-speed integration value is not necessarily added to the low-speed integration value.

  As described above, in this embodiment, high-speed integration is performed when the absolute value of the deviation is between the first and second thresholds, but the high-speed integration is stopped when the absolute value is within the second threshold. The dead zone (within the second threshold range) is set in the control program described in FIG.

<Other examples>
As the web unwinding section continues to operate, the roll diameter changes and the moment of inertia changes. Along with this, the optimum gain of the control system changes. For this, it is desirable to perform index correction or diameter correction described in the background art section. Considering the price of the system, the index correction is cheaper. In the case of radius correction, it is necessary to devise the relationship between the radius and the control parameter. In terms of performance, diameter correction is superior, but index adjustment is easier for parameter adjustment.

  Each parameter adjustment of the controllers 50 and 50A is preferably performed during a trial operation. First, a test operation is performed using a feedback control device that does not use the high-speed integration computing unit 56 of the present invention (the switch SW3 is off), and the parameters of the controllers 50 and 50A are adjusted so that stable control can be performed. Next, it is desirable to perform a trial run using the present invention and adjust each parameter related to the high-speed integration calculator. It is desirable that the parameters related to the high-speed integration arithmetic unit are set to parameters that can be used as they are under general conditions when the controllers 50 and 50A are shipped from the factory. However, it is recommended to check the parameters by trial operation because the characteristics differ depending on the machine to be applied. By using the present invention and the conventional control method in combination, it is possible to establish a control system that achieves both stability and control accuracy.

  The PID calculation unit 55 of the feedback control apparatus of the present embodiment described above literally includes proportional / integral / derivative calculations. However, as this calculation unit, any calculation unit including at least a low-speed integration calculation unit may be used. The present invention can be applied regardless of the type of arithmetic unit. In addition, the integration operation described above includes operations that approximate other integrations such as pulse transfer function operations and linear difference equation operations. As feedback control, robust control that also takes into account modeling errors, adaptive control that changes parameters in response to large fluctuations in the control target, and the like can be applied.

10: Web 20: Roller 30: Intermediate roller 40: Tension detector 50, 50A, 50B, 50C: Controller, 51, 52: Low-pass filter, 53: Signal converter, 54: Subtractor, 55: PID calculation unit, 551 : Proportional calculator, 552: Integral calculator, 553: Differential calculator, 554: Adder, 56: High-speed integration calculator, 57, 58: Adder, 59: Diameter correction unit 60: Electromagnetic brake

To achieve the above object, the feedback control method of the invention according to claim 1 acquires a first calculation result including a low-speed integration calculation result based on a deviation between a target signal and a detection signal to be controlled, In the feedback control method in which the power unit is driven by the first calculation result, the control target is controlled by the power unit, and the control result of the control target is the detection signal, the absolute value of the deviation is set in advance. performing fast integration operation of the deviation when it is within a threshold the second calculation result by adding the integrated value to said first operation result, or high speed with respect to the first operation result An integration operation is performed, and the integration value is added to the first operation result to obtain a second operation result, and the power unit is driven by the second operation result.

According to a sixth aspect of the present invention, there is provided a feedback control device including a first calculation unit including a low-speed integration calculator that performs low-speed integration based on a deviation between a target signal and a detection signal to be controlled, and the first calculation unit. A power unit that is driven by the obtained first calculation result to control the control object; and a detector that detects the control result of the control object, and the signal detected by the detector is the detection signal. In the feedback control apparatus, when the absolute value of the deviation is within a preset first threshold value, a high-speed integration calculation of the deviation is performed, and the integration value is added to the first calculation result to obtain a second value. A high-speed integration computing unit that performs a high-speed integration operation on the first calculation result and adds the integration value to the first calculation result to obtain a second calculation result; The power unit is Characterized by being adapted to the dynamic.

A program according to an eleventh aspect of the invention acquires a first calculation result including a low-speed integration calculation result based on a deviation between an eye signal and a detection signal to be controlled, and drives a power unit based on the first calculation result. In the program for controlling the control object by the power unit and causing the computer to perform an integration operation in a feedback control method using the control result of the control object as the detection signal, the absolute value of the deviation is set in advance. A first step for determining whether or not the difference is within a threshold value, and when it is determined by the first step that the absolute value of the deviation is within a first threshold value set in advance, the fast integration of the deviation second and operation result of the integration value by performing an operation is added to the first operation result, or the first operation result with respect to the integration value by performing high-speed integral operation by the first Calculation Comprising a second step of the second operation result by adding the results, the said power unit by the second operation result is characterized by being so driven.

Claims (15)

A first calculation result including a low-speed integration calculation result is acquired based on a deviation between the target signal and the detection signal of the control target, the power unit is driven by the first calculation result, and the control target is determined by the power unit. In the feedback control method of controlling and using the control result of the controlled object as the detection signal,
When the absolute value of the deviation is within a first threshold value set in advance, a high-speed integration calculation of the deviation is performed and the integration value is added to the first calculation result to obtain a second calculation result, or A feedback control method characterized by performing the high-speed integration operation on the first calculation result, setting the integration value as a second calculation result, and driving the power unit according to the second calculation result. The feedback control method according to claim 1,
When the absolute value of the deviation exceeds the first threshold, the integration value of the high-speed integration calculation is added to or not added to the integration value of the low-speed integration calculation, and the integration value of the high-speed integration calculation is reset to zero And the high-speed integration calculation is stopped. The feedback control method according to claim 1 or 2,
When the deviation can be regarded as zero or zero, the integral value of the fast integral operation is added to the integral value of the slow integral operation, and the integral value of the fast integral operation is reset to zero, and then the fast integral operation is continued. A feedback control method characterized by: The feedback control method according to claim 1 or 2,
When a second threshold value smaller than the first threshold value is provided and the absolute value of the deviation is within the second threshold value, the integration value of the high-speed integration calculation is added or not added to the integration value of the low-speed integration calculation. A feedback control method characterized by stopping the high-speed integration calculation after resetting the integration value of the high-speed integration calculation to zero. In the feedback control method according to claim 1, 2, 3, or 4,
A feedback control method, wherein a maximum integration value of the high-speed integration calculation is set smaller than a maximum integration value of the low-speed integration calculation. Driven by a first calculation unit including a low-speed integration calculator that performs low-speed integration based on a deviation between the target signal and the detection signal to be controlled, and a first calculation result obtained by the first calculation unit. In a feedback control device comprising a power unit for controlling the control object and a detector for detecting a control result of the control object, and using the signal detected by the detector as the detection signal,
When the absolute value of the deviation is within a first threshold value set in advance, a high-speed integration calculation of the deviation is performed and the integration value is added to the first calculation result to obtain a second calculation result, or A high-speed integration arithmetic unit that performs high-speed integration calculation on the first calculation result and uses the integration value as a new second calculation result is provided, and the power unit is driven by the second calculation result. A feedback control device characterized by that. The feedback control device according to claim 6, wherein
When the absolute value of the deviation exceeds the first threshold value, the integral value of the fast integration calculator is added to or not added to the integral value of the slow integration calculator, and the integral value of the fast integration calculator Is reset to zero, and the high-speed integration calculator is stopped. The feedback control device according to claim 6 or 7,
When the deviation can be regarded as zero or zero, the integral value of the fast integration operator is added to the integral value of the slow integration operator, and the integral value of the fast integration operator is reset to zero, and then the fast integration A feedback control device characterized by continuously operating a computing unit. The feedback control device according to claim 6 or 7,
A second threshold value smaller than the first threshold value is provided, and when the absolute value of the deviation is within the second threshold value, the integration value of the fast integration calculator is added to the integration value of the slow integration calculator; and A feedback control device characterized by stopping the operation of the high-speed integration arithmetic unit after resetting the integral value of the high-speed integration arithmetic unit to zero. The feedback control device according to claim 6, 7, 8 or 9,
A feedback control apparatus, wherein a maximum integration value of the high-speed integration calculator is set smaller than a maximum integration value of the low-speed integration calculator. A first calculation result including a low-speed integration calculation result is acquired based on a deviation between the target signal and the detection signal of the control target, the power unit is driven by the first calculation result, and the control target is determined by the power unit. In a program that causes a computer to execute an integration operation in a feedback control method that controls and uses a control result of the control target as the detection signal,
A first step of determining whether the absolute value of the deviation is within a preset first threshold;
When it is determined in the first step that the absolute value of the deviation is within a first threshold value set in advance, a high-speed integration calculation of the deviation is performed and the integration value is added to the first calculation result. A second calculation result, or a second step of performing the fast integration calculation on the first calculation result and setting the integration value as the second calculation result;
With
A program characterized in that the power unit is driven by the second calculation result. The program according to claim 11,
When the absolute value of the deviation exceeds the first threshold, the integration value of the high-speed integration calculation is added to or not added to the integration value of the low-speed integration calculation, and the integration value of the high-speed integration calculation is reset to zero And a third step of stopping the high-speed integration calculation. The program according to claim 11 or 12,
When the deviation can be regarded as zero or zero, the integral value of the fast integral operation is added to the integral value of the slow integral operation, and the integral value of the fast integral operation is reset to zero, and then the fast integral operation is continued. A program comprising the fourth step of: The program according to claim 11 or 12,
A second threshold value smaller than the first threshold value is provided, and when the absolute value of the deviation is within the second threshold value, the integration value of the high-speed integration calculation is added to the integration value of the low-speed integration calculation; A program comprising: a fifth step of stopping the high-speed integration calculation after resetting an integration value of the integration calculation to zero. The program according to claim 11, 12, 13 or 14,
A program characterized in that the maximum integration value of the high-speed integration calculation is set smaller than the maximum integration value of the low-speed integration calculation.

Images