KR102263239B1 - Device for reducing Noise Induced Tracking Error - Google Patents

Abstract

In a preferred embodiment of the present invention, a high pass filter (HPF) is applied to a system output signal including noise fed back to a controller to selectively reduce measurement noise without touching a rapidly changing response component among system output components.

Description

Device for reducing Noise Induced Tracking Error

The present invention relates to a method for reducing measurement noise in a control system.

Recently, as an output error phenomenon due to measurement noise has been discovered in a control system including an actuator, alternatives have been proposed to solve this problem. The output error phenomenon due to measurement noise is a problem not only in proportional integral (PI) control systems but also in systems to which DOB (Disturbance Observer based control) is applied.

In order to solve this problem, a method using LPF (Low Pass Filter) was used in the same manner as in FIG. 1, but in this case, there is still a problem in that even a rapidly changing response component is removed among the output components of the control system or the system to which the DOB is applied. , there is an additional problem that system stability must be considered when using a low pass filter (LPF).

KR 10-1696561

Y. Eun and E. S. Hamby, “Noise induced loss of tracking in systems with saturating actuators and antiwindup,” Journal of Dynamic Systems, Measurement, and Control, vol. 136, no. 5, pp. 054 501-1-6, 2014.

In a preferred embodiment of the present invention, an output error due to measurement noise is reduced in a control system or a system to which DOB is applied.

In a preferred embodiment of the present invention, a high pass filter (HPF) is applied to a system output signal including noise fed back to a controller to selectively reduce measurement noise without touching a rapidly changing response component among system output components.

As a preferred embodiment of the present invention, the measurement noise removing apparatus includes: a proportional integral controller for receiving a difference between an input signal and a feedback signal and performing an integral operation to control the proportional control; an actuator to which an anti-wind-up function is applied and receiving an output signal and a control input signal of the proportional integrator; a plant receiving the output signal of the actuator and performing control; and a measurement noise filtering unit that selectively removes only the measurement noise signal by passing a high pass filter (HPF) after receiving a signal including the output signal of the plant and the measurement noise signal.

In a preferred embodiment of the present invention, the feedback signal of the proportional integral controller is an output value of the measurement noise filtering unit.

As a preferred embodiment of the present invention, the measured noise filtering unit includes: a nominal plant unit for calculating an expected output signal value that is expected to be an output signal value of the plant; and a high-frequency filtering unit that removes the high frequency after removing the expected output signal value calculated by the nominal plant unit from the signal including the output signal of the plant and the measurement noise signal.

As a preferred embodiment of the present invention, a measurement noise removing device for removing measurement noise in a control system provided with a proportional integrator and an anti-windup circuit unit includes an output signal and a measurement noise signal of a plant that is a control object in the control system. a receiver for receiving a feedback signal; and a high-frequency filtering unit that removes a high-frequency component from the feedback signal, wherein an output signal of the filtering unit is provided to the proportional integrator, and the proportional integrator receives a difference between an input signal and an output signal of the filtering unit and is proportional It is characterized in that the integral operation is performed on the control.

As a preferred embodiment of the present invention, the measurement noise removing device further includes a nominal plant unit for calculating an expected output signal value that is expected to be an output signal value of the plant, wherein the high-frequency filtering unit is configured to combine the output signal of the plant with the output signal of the plant. It is characterized in that the high frequency is removed after the expected output signal value calculated by the nominal plant unit is removed from the signal including the measurement noise signal.

In a preferred embodiment of the present invention, the measurement noise removing device has an effect of selectively controlling only the measurement noise while maintaining the response component that changes rapidly among the output of the control system.

As a preferred embodiment of the present invention, the measurement noise removing apparatus has the effect of selectively removing only the measurement noise from the output of the control system and maintaining the rapidly changing response component of the output of the control system by itself.

1 shows an embodiment in which a low pass filter (LPF) is used to reduce distortion due to measurement noise in a control system.
2 to 3 show an example of a control system that selectively filters only measurement noise as a preferred embodiment of the present invention. 2 shows a configuration diagram, and FIG. 3 shows an example of FIG. 2 .
4 shows an example comparing the effects of the measurement noise removing device as a preferred embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 shows an embodiment in which a low pass filter (LPF) is used to reduce distortion due to measurement noise in a control system.

The control system 100 of FIG. 1 includes a proportional integral controller 110 , an actuator 120 , a plant 230 , an anti-windup circuit unit 140 , and an LPF 150 . 1, r(S110) is a reference signal, u(S112) is a control input, v(S120) is an actuator output signal, y(S130) is a control system output signal, and n(S140) is measurement noise, respectively.

로 표시하며, α는 액츄에이터의 최소 한계값, β는 액츄에이터의 최대 한계값을 나타내며, 액츄에이터(120)는 아래와 같이 정의된다. The actuator 120 is

, α is the minimum limit value of the actuator, β is the maximum limit value of the actuator, and the actuator 120 is defined as follows.

In the control system 100 of FIG. 1 , a feedback signal y+n ( S142 ) including measurement noise is input to the LPF 150 , and provides a signal with reduced measurement noise to the proportional integral controller 110 . However, in this case, there is a problem affecting the response of the system because the rapidly changing response component itself present in the output signal y (S130) of the control system also passes through the LPF 150 and is filtered.

2 to 3 show an example of a control system that selectively filters only measurement noise as a preferred embodiment of the present invention. 2 shows a configuration diagram, and FIG. 3 shows an example of FIG. 2 .

The control system 200 shown in FIGS. 2 to 3 includes a proportional integral controller 210 , an actuator 220 , a plant P(s) 230 , an anti-windup circuit unit 240 , and a measurement noise filtering unit 250 . do. Also, r(S210) is a reference signal, u(S212) is a control input, v(S220) is an actuator output signal, y(S230) is a control system output signal, and n(S240) is measurement noise, respectively. And, Kaw represents the anti-windup gain.

The control system 200 shown in FIGS. 2 to 3 is a proportional integral control system to which the anti-wide-up technique is applied. The control system 200 receives the reference signal r (S210) and the filtered feedback signal y' (S254) from the proportional integral controller 210, and performs an integral operation for proportional control to perform control.

After that, the output value of the proportional integral controller 210 and the control input u (S212) are summed and input as an input signal of the actuator 220, and the output value v (S220) of the actuator is input to the plant 230, and the plant 230 generates an output value y (S230). In a preferred embodiment of the present invention, the plant 230 includes all objects to be controlled.

The actuator 220 implements the anti-wind-up function through the anti-wind-up circuit unit 240 . When the output of the proportional integral controller 210 is saturated, an overshoot increases in the speed response characteristic of the actuator 220, and a phenomenon in which the time response becomes slow occurs. It is said In a preferred embodiment of the present invention, an anti-windup circuit unit 240 is implemented in the actuator 220 to prevent such a windup phenomenon from occurring.

Specifically, for the difference between the output value v (S220) of the actuator 200 and the control input value u (S212) of the actuator 200, the signal amplified by the anti-windup gain value Kaw is again amplified as the control input value of the actuator 200 It is implemented to sum to u(S212). Thus, it is possible to remove the windup occurring in the actuator 220 .

The measurement noise filtering unit 250 is implemented to selectively filter the measurement noise n(S240) from the feedback signal in which the measurement noise n(S240) is added to the output value y(S230) of the plant 230 . As a preferred embodiment of the present invention, the measurement noise filtering unit 250 selectively filters only the measurement noise n (S240) by using a high pass filter (HPF) 270 , and the output value y ( S230 ) of the plant 230 . ) can be preserved without filtering the rapidly changing response components included in the

Referring to FIG. 3 , the measurement noise filtering unit 250 selectively removes only the measurement noise from the output signal of the plant including the measurement noise. The measurement noise filtering unit 250 includes a nominal plant unit 260 and a high frequency filtering unit 270 .

The nominal plant unit 260 calculates an expected output value to be output from the plant 230 . After receiving the output signal y+n of the plant including the measurement noise, the measurement noise filtering unit 250 outputs y1 (S270) obtained by subtracting the expected output value of the plant 230 calculated by the nominal plant unit 260. do.

It can be expressed as y1=[P(s)-Pn(s)]v+n, where Pn(s) represents a nominal plant. The measurement noise filtering unit 250 inputs the y1 signal S270 to the high frequency filtering unit 280 . The high-frequency filtering unit 270 may be implemented in the form of 1-Q(s), and in this case, Q(s) represents the LPF.

The high-frequency filtering unit 270 does not pass the [P(s)-Pn(s)]v part, which is a low-frequency component in the y1 signal, but only the measurement noise component. After passing only the measurement noise component, the measurement noise filtering unit 250 may transmit only the y signal as a feedback signal to the proportional controller 210 by differentiating it from the y+n signal.

4 shows an example comparing the effects of the measurement noise canceling device as a preferred embodiment of the present invention.

4, 410 is a reference signal indicating the speed of the motor. 420 denotes an output value when no filtering is applied to remove output noise. 430 shows an example in which low-frequency filtering is applied as shown in FIG. 1 . Reference numeral 440 shows an example in which high-frequency filtering is applied as a preferred embodiment of the present invention.

So far, with respect to the present invention, the preferred embodiments have been looked at. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments are to be considered in an illustrative rather than a restrictive sense. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

Claims (5)

a proportional integral controller for receiving a difference between an input signal and a feedback signal and performing an integral operation for proportional control;
an actuator to which an anti-windup function is applied and receiving an output signal and a control input signal of the proportional integral controller;
A plant that receives the output signal of the actuator and performs control; And
Measurement noise removal comprising a; a measurement noise filtering unit that selectively removes only the measurement noise signal by passing a high pass filter (HPF) after receiving a signal including the output signal of the plant and the measurement noise signal Device. The method of claim 1, wherein the proportional integral controller
The feedback signal is an output value of the measurement noise filtering unit. The method of claim 1, wherein the measurement noise filtering unit
a nominal plant unit for calculating an expected output signal value that is expected to be an output signal value of the plant;
and a high-frequency filtering unit that filters the high frequency after removing the expected output signal value calculated by the nominal plant unit from the output signal of the plant to which the measurement noise is added, and in this case, the high-frequency filtering unit uses only the measurement noise This is filtered, and the difference value between the output signal value of the plant which is a low frequency component and the expected output signal value is not filtered. A measurement noise removal device for removing measurement noise from a control system having a proportional integral controller and an anti-windup circuit unit, comprising:
a receiver for receiving a feedback signal including an output signal and a measurement noise signal of a plant to be controlled by the control system; and
and a high-frequency filtering unit that removes a high-frequency component from the feedback signal, wherein an output signal of the filtering unit is provided to the proportional integral controller, the proportional integral controller receiving a difference between an input signal and an output signal of the filtering unit Measurement noise removal device, characterized in that performing an integral operation for proportional control. 5. The method of claim 4,
Further comprising; a nominal plant unit for calculating an expected output signal value of the expected output signal value of the plant, the high-frequency filtering unit
Measurement noise removing device, characterized in that the high frequency is filtered after removing the expected output signal value calculated by the nominal plant unit from the signal including the output signal of the plant and the measurement noise signal.

Images