WO2002048806A1

WO2002048806A1 - Feedback control device

Info

Publication number: WO2002048806A1
Application number: PCT/JP2001/010951
Authority: WO
Inventors: Wennong Zhang; Yasuhiko Kaku
Original assignee: Kabushiki Kaisha Yaskawa Denki
Priority date: 2000-12-14
Filing date: 2001-12-13
Publication date: 2002-06-20
Also published as: TW569084B

Abstract

A feedback control device comprising a controller, a controlled object controlled by the controller, and an observer which receives the control output from the object and the output of the controller and makes the controlled model output a feedback signal. This observer and an observer compensator which receives the difference between the control output and the output of an element model can constitute a control system of excellent responsiveness by inputting the sum of the output of the observer compensator and the output of the controller in the controlled model, easily realizing a stable observer.

Description

Description Feedback control device Technical field

The present invention relates to a feedback control method and apparatus in which a delay element exists in a controller or a controlled object, or a dead time exists in input / output of the controlled object. Background art

In most control systems such as servo and process control, there is a delay element such as dead time / fill time for the controller or control target. Fig. 3 is a block diagram of an ordinary feedback control system.

In FIG. 3, 2 is a main controller (for example, a PID controller), 3 is a delay element of the controller, 4 is a delay element of the control target, and 5 is an element having no delay of the control target. In such a control system, since the phase of the delay element is delayed, sufficient gain cannot be obtained because the gain of the controller cannot be increased. Here, compensation control for phase lag is required.

Conventionally, it is often used to add a phase lead compensating element 12 in series as shown in FIG. 4 to compensate for phase lag. T _{_a>} T T _a as _b, lever to appropriately set the T _b, becomes the phase lead compensator 1 2 phase is advanced, that is increasing the gain of the main control unit 2, to increase the control performance it can. .

However, in the above-described conventional phase compensation control method, the gain of the phase lead compensating element in a high frequency range becomes large, so that there is a problem that high-frequency oscillation is likely to occur.

In addition, most control systems such as support systems and process control have dead time for input or output. Fig. 8 is a block diagram of a conventional feedback control system. In FIG. 8, reference numeral 22 denotes a normal controller (for example, a PID controller), and reference numeral 23 denotes a control object including dead time. In such a control system, since the phase of the dead time element is delayed, the gain of the controller cannot be increased, and sufficient response characteristics cannot be obtained. Here, compensation control for the dead time is required.

Conventionally, a Smith compensator as shown in Fig. 9 is often used to compensate for dead time. In FIG. 9, reference numeral 25 denotes a prediction model of a control target, and reference numeral 26 denotes a dead time element. Focusing on the control input and feedback signal, the control system in Fig. 9 is equivalent. It can be rewritten as shown in Figure 6. According to Fig. 6, the stability of the feedback system is the same as that of the system without dead time, the gain of the controller C (s) can be increased, and the control output y can accurately follow the target input r.

Let us consider the disturbance elimination characteristics of the conventional Smith method. Assuming that a disturbance d exists at the control input end as shown in Fig. 10, the transfer function from the disturbance d to the control output y is

It is given by d (s) l + C (s) P (s) l + C (s) P (s) ^LJ . Assuming that the steady-state value of the control output y for the step disturbance d (s) = lZs is y _sd , = lim = lim

^{^{= l 1 ifim5 1 il + C}} (5) P (5) l + C (s () P (s) L "J, -. J J ... (2) become long with C (s) is the integrator , Y _si = lim ^)-^) ^e " ^is J = ^ lim ^) · '(3). If P (s) has a pole at s = 0, then y _sd ≠ 0. That is, in the Smith method, there is a problem that a steady-state error is generated for a controlled object having a pole of s = 0. Also, from Eq. (1), if P (s) is unstable, there is a problem that the output diverges regardless of any small disturbance.

An object of the present invention is to provide a control device which increases a control gain and has good responsiveness.

Another object of the present invention is to compensate for the phase delay of a delay element in a control system having a delay element in a controller or a controlled object so as not to cause high-frequency vibration, and to accurately control a control output to a target input. The objective is to provide a feedback control device that can follow well.

Still another object of the present invention is to provide a dead-time compensated feedback control in which a steady-state deviation does not occur for a control target having a pole of s = 0 and a stable control system can be configured for an unstable control target. It is to provide a device. Disclosure of the invention

In order to achieve the above object, the present invention provides a controller, a control target controlled by the controller, a control output from the control target and an output of the controller, and a control target model output. In a feedback control device including an observer serving as a feedback signal, the observer includes: an observer compensator that inputs a difference between the control output and an output of an element model; an output of the observer compensator; and the controller The control output follows the target input for the control system in which a delay element exists in the controller or the control target. In a control device for performing feedback control, a difference between a control output and an output of an observer is input to a compensator of the observer, and the observer The output of the compensator and the output of the main controller are added to the model of the element having no delay of the control target, and the output of the model of the element having no delay of the control target is output as the model of the delay element of the controller. And an observer configured to use a signal passed through the model of the delay element of the controlled object as an output of the observer, and subtracting an output of the model of the element having no delay of the controlled object from the target input to reduce the main output. A controller that inputs the output of the main controller to a delay element of the controller as a control input, and controls a control target. In a feedback control device in which a control output is made to follow a target input for a control target having time, the difference between the control output and the output of a dead time element of the control target is determined by an error. Input to the compensator of the buzzer, add the output of the compensator of the observer and the control input, and input the result to the prediction model of the controlled object, and output the output of the predicted model of the controlled object. The observer is configured to input the predicted value of the control output from the target input and input to the controller, and the output of the controller is output to the controller. It is characterized as a control input. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a block diagram showing a configuration principle of a first control system of the present invention, FIG. 2 is an equivalent block diagram of FIG. 1, FIG. 3 is a block diagram of a normal feedback control system, FIG. 4 is a block diagram of a conventional phase lead compensation control system, FIG. 5 is a block diagram showing a configuration principle of a second control system of the present invention, and FIG. 6 is an equivalent block diagram of FIGS. Fig. 7 is a block diagram of Fig. 5 when disturbance is considered, Fig. 8 is a block diagram of a conventional feedback control system, and Fig. 9 is a Smith compensator. FIG. 10 is a block diagram of the control system, and FIG. 10 is a block diagram when disturbance is considered in FIG. '' · Best mode for carrying out the invention

An embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration principle of the control system of the present invention. In Fig. 1, 2 is the main controller, 3 is the delay element of the controller, 4 is the delay element of the control target, 5 is the element with no delay of the control target, 7 is the model of the element with no delay of the control target, 8 Is the model of the delay element of the controller, 9 is the model of the delay element to be controlled, and 11 is the compensator of the observer.

Generally, it is necessary to insert a filter to remove noise when constructing a controller. In addition, there are dead time elements and first-order lag elements in the control input or control output of the control target. Unless these delay elements are separated, it is difficult to configure the main controller to satisfy both stability and control performance at the same time. Therefore, the controller is divided into the main controller 2 and the controller delay element 3, and an observer including the controller delay element is constructed based on the main controller output ui and the control output y, and the observer delay An output having no is referred to as a feedback signal y _f . More specifically, the output u of the main controller 2 and the output of the compensator 11 of the observer are added, and the result is input to the model 7 of the element having no delay of the control target. The output of the model 7 delay of the control target is no element is a feedback signal y _f of the control system is rectangular, inputs a signal obtained by subtracting the feedback signal y _f subtractor 1 from the target input r to Mein controller 2, The output u of the main controller 2 is passed through the delay element 3 of the controller, and then input to the control target to control the control target. On the other hand, the model 8 of the delay element of the controller and the delay The signal subtracted from the control output y is input to the observer compensator 11 through the model 9 and input to the subtractor 10.

From Fig. 1, in the open loop of the control system, the transfer function from the output of the main controller 2 to the feedback signal y _£ is

^-P (s) · · · · (4) ', so the control system in Fig. 1 can be rewritten equivalently as in Fig. 2. According to FIG. 2, the stability of the feedback control system is the same as that of the system without delay, and the gain of the main controller 2 can be increased. The transfer function from the target input r to the control output y is

Therefore, the output y can accurately follow the target input r by increasing the gain of the main controller 2.

Although the observer loop has a delay element, the input / output characteristics are not related to the compensator 11 of the observer from Equation 5, so the gain of the compensator 11 of the observer may be set low. Therefore, a stable observer can be easily configured. Another embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a block diagram showing the configuration principle of the second control system of the present invention. In FIG. 5, 22 is a controller, 23 is a controlled object including dead time, 25 is a prediction model of the controlled object, 26 is a dead time element, and 28 is a compensator of the observer 20.

5 that the transfer function from the control input u to the feedback signal y _f

^ = () · · · (6)

s), the control system in Fig. 5 can be rewritten equivalently as in Fig. 6. According to Fig. 5, the stability of the feedback system is the same as that of the system without dead time, and the gain of the controller C (s) can be increased. The transfer function from the target input r to the control output y is

As) C (s) P (s) e- ^Ls … (7)

φ) l + C (s) P (s), so the output y can accurately follow the target input r by increasing the gain of the controller C (s).

In order to consider the disturbance rejection characteristics, a disturbance d is inserted at the control input terminal as shown in Fig.7. From Fig. 7, the transfer function from the disturbance d to the control output y is

) '(P (s) e- ^Ls _i C (s) P (s). P ( _S ) e ~ ^Ls … (8) d (s) l + C (s) P (s) l + C (s ) P (s) l + C ₀ (s) P (s) e- ^Ls

Given by If the steady-state value of the output for the step disturbance d (s) = lZs is y _sd , ^; _irf = lim.v (= lim¾ ^; ( ⁵ )

Becomes If both the controller C (s) and the observer compensator (:. (S) have an integrator,

y _Sil = o · · · do). In other words, no steady-state deviation occurs even for a controlled object with a pole of s = 0. Also, from equation (6), even if P (s) is unstable, C (s) replaces P (s) with C. If (s) stabilizes P (s) e, the output does not diverge. From equation (7), the input / output characteristics are C. Because it does not related to the (s), P (s) e- Ls stable to. Ru by sea urchin C. The gain of (s) can be set low.

As described above, the present invention comprises an observer including a delay element of a controller, and an output without delay of an observer is used as a feedback signal, so that the control system includes a feedback control section and an observer section. Is separated into The stability of the feedback loop is the same as that of a system without delay, and the control gain can be increased, so that the response performance of the control system can be improved. On the other hand, the observer loop has a delay element, but since the observer's compensator does not affect the input / output characteristics, the gain of the observer's compensator may be set low, so a stable observer can be easily configured. it can.

Further, the present invention constitutes an observer of a dead time system and uses a predicted value of a control output as a feedback signal, so that the feedback control part and the observer part are separated, and the stability of the feedback loop has no dead time. It becomes the same as the system and the control gain can be raised, so the response performance of the control system can be improved. In addition, because it has an observer, it is not only stable but also resistant to disturbances. Industrial applicability

According to the present invention, a control system having good response performance can be configured, and a stable observer can be easily configured.

Claims

The scope of the claims

1. A feedback comprising a controller, a controlled object controlled by the controller, a control output from the controlled object and an output of the controller, and an observer using the output of the controlled object model as a feedback signal. In the control device, the observer may include: an observer compensator configured to input a difference between the control output and the output of the element model; and a sum of an output of the observer compensator and an output from the controller. Feedback control characterized by input to the model

2. In a control device that performs feedback control that causes a control output to follow a target input in a control system in which a delay element exists in a controller or a control target, a difference between the control output and the output of the observer is used as a compensator of the observer. 1 Input to 1 and add the output of the observer's compensator 1 1 and the output of the main controller 2 to the model 7 of the element with no delay of the controlled object. An observer is configured so that a signal obtained by passing the output of the model 7 through the delay element model 8 of the controller and the control target delay element model 9 is used as the output of the observer, and the control is performed from the target input. The output of the model 7 of the element having no delay is subtracted and input to the main controller 2, and a signal obtained by passing the output of the main controller 2 to the delay element 3 of the controller is used as a control input. Feedback controller you and controlling the controlled object.

3. In a feedback control device in which a control output follows a target input for a control target having a dead time in input or output, a difference between the control output and the output of a dead time element of the control target is provided. Is input to the compensator of the observer, the output of the compensator of the observer and the control input are added and input to the prediction model of the control target, and the output of the prediction model of the control target is used. The observer is configured to input to the dead time element and, on the other hand, to obtain the predicted value of the control output, to subtract the predicted value of the control output from the target input and input to the controller, and output the output of the controller. Is the control input.