JPH04203601A - Oscillation preventing method for automatic control valve with electropneumatic positioner - Google Patents

Abstract

PURPOSE:To stabilize opening-closing control of a valve by connecting a compensation circuit having specified transmission functions in parallel between the input and output sides of a controller, thereby preventing the opening-closing oscillations of a drive valve in the case a pneumatic pressure tubular passage is made long. CONSTITUTION:A compensation circuit H' by the Smith method is connected between the input and output sides of a controller 3, and thereby the opening- closing oscillations of a drive valve 1 in the case a pneumatic pressure tubular passage 8 is made long. By the use of two diaphragm drive valves 1 each having a electropneumatic converter 2, the transmission function H of the circuit H' is made the same as the object to be controlled. This function H is set to satisfy the formula I. A feedback signal 9 of each drive valve 1 is input to a computation circut Hc, and the differential signal is computed to be fed back to the controller 3.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention relates to an improvement in the control method of an automatic control valve equipped with an electropneumatic positioner, and the invention relates to an improvement in the control method of an automatic control valve equipped with an electropneumatic positioner. Stable opening/closing control is now possible without causing opening/closing vibration. The present invention relates to a method of controlling an automatic control valve equipped with an electropneumatic positioner.

(Prior art) An automatic control valve equipped with a so-called electropneumatic positioner usually has a 11th
As shown in the figure, it is comprised of a drive valve 1 equipped with a diaphragm drive section 1a and a linear potentiometer 1b, an electropneumatic converter 2 equipped with a flapper 2a made of a piezoelectric element, a PID controller 3, etc.

By inputting the input operation signal 4 to the PID controller 3, a control signal 5 is applied to the flapper 2a.

The flapper 2a is operated in a direction to close the nozzle 2b.

That is, when the flapper 2a operates, the pressure in the pressure chamber 6a of the pilot valve 6 increases as shown in FIG. 12, and the balance with the pressure in the back pressure chamber 6b is disrupted, causing the spool 6c to rise. As a result, the port 6d is opened and the driving air 7 is applied to the diaphragm driving section 1a through the pneumatic pipe line 8, and the stem 1c is displaced.

The displacement of the stem 1c is converted into an electric signal by the linear potentiometer 1b, and the converted displacement signal (feedback signal) 9 is fed back to the controller 3, whereby the control signal 5 changes and the stem of the driven valve 1 The correction operation is performed until the IC is displaced by an amount commensurate with the input operation signal 4, and a stem displacement of the drive valve 1 that is always proportional to the input operation signal 4 is obtained.

(Problem to be Solved by the Invention) Therefore, in the conventional automatic control valve having the configuration as shown in FIG. , and the distance of the pneumatic pipe line 8 is short,
Stem displacement proportional to the input operation signal 4 can be obtained relatively quickly and accurately.

On the other hand, if the automatic control valve is to be used in an atmosphere of flammable gas, the control parts such as the electro-pneumatic converter 2 and the control circuit 3 should be separated from the drive valve 1 for safety reasons. There is a need.

However, if the control section and the drive valve 1 are separated, the pneumatic pipe line 8 will inevitably become longer, resulting in a so-called transmission delay of the pneumatic signal. As a result, the drive valve 1 causes oscillation (or hunting) as shown in FIG. 9, which will be described later, and the control operation becomes extremely unstable.

The present invention aims to solve the above-mentioned problem in an automatic control valve equipped with an electropneumatic positioner, that is, the problem of oscillation of the driven valve 1 when the pneumatic pipe line 8 becomes long.
A method for preventing oscillation of an automatic control valve that enables stable valve opening/closing control without causing oscillation by adjusting the compensation circuit provided in the PID controller 3 even if the pneumatic pipe line 8 is long. This is what we provide.

(Means for Solving the Problem) The inventor of the present invention previously aimed at improving the dynamic characteristics of an automatic control valve equipped with this type of electropneumatic positioner.

In addition to developing a relatively simple calculation method for the transmission delay of the air signal, we also analyzed the effect of the length of the pneumatic pipeline on the transmission time delay of the air signal and published this [Proceedings of the Japan Society of Mechanical Engineers] (Part B) Volume 55, No. 512 (1989-
April)].

In addition, the inventor of the present invention previously conducted various operation tests aimed at stabilizing this type of automatic control valve using an automatic control valve response experiment device as shown in Fig. 13, which is necessary for analyzing the dynamic characteristics. The displacement of the flapper 2a, the displacement of the stem 1c, the output pressure of the electro-pneumatic converter 2, etc. are measured to identify the parameters of each element constituting the response experimental device, and the electro-pneumatic converter 2 having a nozzle flapper mechanism is is a second-order lag element, and the drive valve 1 with the diaphragm drive section 1a is a first-order lag element.

Assuming that each of the components of the pneumatic equipment for driving the valve has linear characteristics, a mathematical model representing the dynamic behavior of the driving valve 1 was devised to be represented by a block diagram as shown in FIG. 14.

In Fig. 14, K: Gain of the entire control system KC Gain of the two control circuits v/m AKp: Proportional gain of the PID controller Ks: Gain of signal conversion v/m AKV: Gain of the diaphragm driven valve mm /PaL: Dead time S S Niraplus operator TV: Time constant of diaphragm driven valve SRo: Operation signal mA Y: Displacement of driven valve mm α: Gain of linear potentiometer v/mmωn:
Natural frequency rad/sζn of the nozzle flapper: This is the damping rate of the nozzle flapper, and each value was obtained from a test using the response experiment apparatus shown in FIG. 13, and has the following values.

Kc=0.85v/mA, Kv=0.87mm/P
a.

Ro=20mA, ωn=3.52rad/s.

Tv=o, 1s, α=1.33v/mm, ζn=
0.67, L=0.05s, 0.30s In addition, the dead time L=0.05s (when L=0.3m) and L=0.
30s (when L=30m) is calculated by the calculation method disclosed in the aforementioned Transactions of the Japan Society of Mechanical Engineers (B edition), Vol. 55, No. 512.

In addition, we conducted a simulation experiment using the block diagram of an automatic control valve with an electropneumatic positioner shown in Figure 14, and the results showed that: ■ The transmission delay of the pneumatic signal and the gain of the entire control system have a large influence on the occurrence of oscillation phenomena. (1) stability limits of the control system, and (2) conditions for the occurrence of oscillation phenomena caused by transmission delay time (dead time).

Furthermore, from the analysis of the results of the simulation experiment, the present inventor has determined that: ■ The modeling using the block diagram shown in FIG. 14 is appropriate; ■ The controller 3 It is necessary to conduct a quantitative study that takes into account nonlinear elements such as the seal part of the drive valve 1, etc., and ■To prevent the oscillation phenomenon.

They have learned that not only a PID control circuit but also other compensation circuits are required, and have made this public.

November 1984.

Therefore, regarding the prevention of oscillation phenomena in the control system, ○, J,
A stabilization method (hereinafter referred to as the Smith method) proposed by Mr. M and S+aith is known.

In other words, the Smith method stabilizes the system with dead time by incorporating a model of the controlled object into the controller 3, eliminating the dead time from the closed loop, and reducing the problem to a normal finite-dimensional control problem. It is. In other words, this method is a method of configuring a transfer function G p (s)-e-SL bag circuit for a controlled object having dead time. Therefore, the response in this case only needs to take into account the dead time.

Let Gp(s) be the transfer function of the remaining controlled object excluding dead time. The block diagram of the automatic control valve equipped with the electropneumatic positioner at this time is expressed as the above-mentioned No. 151iii1. However, it is assumed that the transfer functions of each element are all linear systems.

Here, the transfer function G(s) between the target value and the controlled variable is defined as Gc(s), where Gc(s) is the transfer function of the controller.

In such a case, no matter what compensation element is introduced, there is a time delay of L between the target value of the control system and the controlled variable. Therefore, if a control system having a closed circuit transfer function in place of Equation 1 (1) can be realized, this control system is considered to be optimal for a controlled object having dead time.

Next, let us consider configuring a control system having a closed circuit transfer function as shown in equation (2).

For this purpose, equation (2) is transformed into the following equation 6, that is, equation (3
) on the right side of

to be accomplished.

The following relationship can be obtained from equations (3) and (4). Furthermore, a configuration as shown in FIG. 16 can be obtained to realize the closed circuit transfer function of equation (5). This is OlJ, M, Sm1t
This is a control system proposed by H.

The closed circuit transfer function of the control system in Fig. 16 is of course expressed by equation (2), but as is clear from equation (2), a pure dead time element with a transfer function of e- is outside the feedback circuit. , Gc(s) can be designed without regard to dead time in a control system as shown in FIG.

As a result, as can be seen from Fig. 17, the calculation of the response to the input of the control system of O, J, M, and S+with is as follows:
Only the dead time needs to be considered.

The present invention was created based on various operational tests and analyzes of automatic control valves equipped with electropneumatic positioners as described above, and the flapper 2a made of a piezoelectric element is activated by the control signal 5 from the controller 3. The equipped electro-pneumatic converter 2 is actuated to supply driving air pressure to the driven valve 1 through the pneumatic line 8, and at the same time, the displacement signal 9 proportional to the displacement of the stem 1c of the driven valve 1 detected by the linear potentiometer 1b is transmitted to the driven valve 1. In an automatic control valve equipped with an electro-pneumatic positioner that maintains the displacement of the stem 1c at a position corresponding to an input electrical operation signal 4 to the controller 3 by foot-hacking the controller 3, the controller 3 The transfer function L Gf(s)・Gp(s)(1-e-) (However, Gf(
s) Transfer function of the beam potentiometer 1b, Gp(
S) is the transfer function of the controlled object when there is no dead time, Gp
By connecting a compensation circuit H' having a transfer function of a controlled object with dead time], opening/closing vibration of the drive valve 1 is prevented when the pneumatic pipe line 8 is lengthened. This is the basic structure of the invention.

(Function) L By connecting the compensation circuit H' having the transfer function 0f(s)・Gp(s)(1-e-) in parallel between the input and output of the controller 3, the dead time element is reduced by the controller. Therefore, the transfer function Gc(s) as the controller 3 has no relation to the dead time.

As a result, even if the pneumatic pipe line 8 is long and the transmission delay of the pneumatic signal (dead time opening L) becomes large, the response to the input/output of the controller 3 becomes irrelevant to the dead time, and the control system becomes stable. oscillation is effectively prevented.

(Example) Hereinafter, the present invention will be described in detail based on the drawings.

The basic configuration of an automatic control valve equipped with an electropneumatic positioner according to the present invention is almost the same as that shown in FIG. These are the points that are provided.

FIG. 1 also shows the nozzle flapper mechanism 2a shown in FIG.
, 2b is assumed to be a second-order delay element, the diaphragm-driven valve 1 is assumed to be a first-order delay element, and the transmission delay of the pipe line 8 is assumed to be a dead time. It is a block diagram of the control system of an automatic control valve with a pneumatic positioner, where A is the transfer function of the control 3, and B is the electropneumatic converter 2.
C is the delay dead time, D is the transfer function of the drive valve 1, E is the saturated nonlinear element, and F is the transfer function of the linear potentiometer.

In Fig. 1, KP: Proportional gain of control circuit v/n+AKv: Gain of diaphragm-driven valve mm/paL = Dead time of controlled object S m 1 Linear coefficient of nonlinear element RO: Input electrical operation signal IIA S Nira plus operator T1: Integral time of control circuit S Td: Differential time of control circuit S Tv: Time constant of diaphragm driven valve SY: Drive valve displacement am ωn: Natural frequency of electropneumatic converter rad/s2: To nonlinear element input amplitude V α : linear potentiometer gain v/rnIIζ
n: Attenuation rate δ of the electro-pneumatic converter: Saturation value V of the input of the nonlinear element.

In the first @, the PID controller (controller) 3
The transfer function Gc(s) is: Gc(s)=Kp(1+-7-+Tds)
-(2-1) Give as IS. The transfer function of linear potentiometer b is Gf(s)=α. From these electro-pneumatic converter 2
The transfer function Gρ(S) of the controlled object is obtained by removing the dead time from the transfer function Gn(s) of the diaphragm-driven valve 1 and the transfer function Gy(s) of the diaphragm-driven valve 1, as follows:
...(2-2).

On the other hand, since automatic control valves produce oscillation phenomena (limit cycles), the system is considered to include nonlinear elements. Therefore, it is assumed that a saturated nonlinear element E shown in FIG. 2 exists in the flapper 2a made of a piezoelectric element of the electropneumatic converter 2.

Incidentally, next to the saturated nonlinear element E as shown in FIG. 2, a compensation circuit based on the entry error method is applied to the control system of an automatic control valve with an electropneumatic positioner as shown in FIG.

The Smith method stabilizes the dead time system by incorporating a model of the controlled object into the controller 3 to eliminate dead time from the closed loop and reducing the problem to a normal finite-dimensional control problem. That is, in this method, the dead time is also L. By identifying the transfer function Gp(s)-e- of the controlled object, Gp(s) for Gc(s) of the controller 3 is determined.
This is a method of configuring a local feedback circuit of -(1-Le). Therefore, in response in this case, only the dead time needs to be considered. However, Gp(s) is the transfer function of the remaining controlled object excluding the dead time.

FIG. 3 shows a control system for an automatic control valve with an electropneumatic positioner in consideration of the compensation circuit based on the entry error method.From FIG. 3, the characteristic equation of the system is as follows.

1 +N(z)Gc(s)Gp(s)Gf(s)” O
・-・(3-1) Here, as s=jω, equation (3-1)
When rearranged, -one side = Gs (jω) ・
...(3-2). However, Gs(s)=Gc(
s)Gp(s)Gf(s).

Further, from FIG. 3, the displacement of the drive valve 1 when a step-like electrical operation signal is input to the system is determined by numerical calculation. From FIG. 3, assuming input R(s) and output Y(s), the following differential equation can be obtained from the transfer function between input and output.

However, when u(.) is a unit step function, each coefficient b-bG is as shown in the following equation.

b E = Tv, b 2 = 1
+ 2ζnωnTvb3=2ζnωn+ωn2Tv,
b4=ωn2bs ” K , b s =K Ro /
Fig. 4 is the block diagram of Fig. 3 above.

This shows the transient characteristics of the driven valve 1 by numerical calculation using Equation 3-4 when a step-like human force 4 is applied.In addition, each parameter necessary for the analysis of the dynamic characteristics is obtained from response experiments of each component of the system. The following values obtained are used. Also,
The dead time is set as L=0°4O8 (length of pipe 8 is 40 m).

Kp=1.70 v/+aA, Ti=oos,
Td=OsRo=20mA, ωn=2. orad
/sy ζn=o, 67Kv=0.87 +u+/p
a, α=1.33 v/m+s, Tv=O,l
sNext, in order to compare the results of the theoretical analysis based on the block diagram shown in Fig. 3 with the actual operation of the automatic control valve device, an automatic The dynamic characteristics of the control valve were measured.

In addition, in FIG. 13, 11 is a pressure gauge, 12 is a regulator, and 13 is a compressor.

When a step-like input of 20@A is given to the input side child as the input operation signal 4, the length of the pipe line 8 is set as Q=0.3 and u=
FIG. 5 shows the transient characteristics of the driven valve 1 when the length is 40 m. From Fig. 5, when the length of the pipe line 8 is Q-0.3 m, the displacement of the single-hair valve 1 is stable, whereas Q=
In the case of 40 m, the driving valve 1 has a period of 2.04 s and an amplitude of 2.04 s.
It was found that an oscillation phenomenon of 89 m was produced.

Incidentally, the displacement of the driving valve 1 is obtained by recording the electric signal 9 detected by the potentiometer 1b in real time using a multi-pen recorder.

Further, from the results of an operation test using the experimental apparatus shown in FIG. 13, it was found that the oscillation phenomenon was caused by the gain of the control circuit 3 and the length of the conduit 8.

FIG. 6 shows the experimental results in which the length of the conduit 8 is plotted on the vertical axis and the gain of the control circuit 3 is plotted on the horizontal axis, and an oscillation phenomenon was confirmed in the shaded area. Broken line a indicates that there is no hysteresis in drive valve 1 and the rise time (valve opening 90%) is 0.
．． The minimum gain for 1 s is shown. Further, the circles in FIG. 6 indicate the limit points of oscillation determined through experiments. Note that in the region to the left of the broken line a, although the operation is somewhat stable, the characteristics have hysteresis and are not within the practical range. Therefore, from FIG. 6, the range in which the driven valve 1 operates stably is the area between the broken line a and the actual &b.

It has also been found that even in this region, when the length of the pipe exceeds 40 m, oscillation occurs, and measures to prevent it are necessary.

Furthermore, from the results of the operation test using the experimental apparatus shown in FIG. 13, the occurrence of oscillation in the drive valve 1 is considered to be due to some nonlinear element in the system. Therefore,
Here, when the correlation between the input voltage 4 to the control circuit 3 and the applied voltage 5 to the piezoelectric element was investigated, it was found that there is a saturation characteristic between the two as shown in FIG.

In other words, the input voltage 4 is linear in the range of +1.0 to 4.6V, and the displacement of the driven valve also changes linearly in the linear portion of the input voltage, but at other times there is a saturation point, and this is an oscillation phenomenon. was found to be the cause.

Next, the compensation circuit H using the Smith method, which is the main part of the present invention.
A control system for an automatic control valve with an electropneumatic positioner including the following will be explained.

First, a model of the compensation circuit H' was constructed, and a control system using the model was tested to see if it could effectively prevent the oscillation phenomenon.

FIG. 8 is a block diagram of the control system, and the part surrounded by the dotted line shows the compensation circuit H' based on the Smith method. When configuring the compensation circuit H', its transfer function H is made to be the same as that of the controlled object.

Two diaphragm-driven valves 1 equipped with electro-pneumatic converters 2 are used.

That is, the compensation circuit H' has two dead times L = OCs] (pneumatic pipe Q = 0.3 [m] and dead time L = 0.3 [s] (pneumatic pipe Q = 30 [m]). It consists of a diaphragm driven valve 1 equipped with an electro-pneumatic converter 2, each of which has a configuration in which nine feedback signals 9 are calculated by a subtraction circuit He, and the difference signal is fed back to a control circuit 3. .

When operating the control system shown in FIG. 8, since the characteristics of the piezoelectric element 2a of the electropneumatic converter 2 of each drive valve 1, the linear potentiometer 1b, etc. are different, the same actual system as the controlled object is applied to the compensation circuit H'. In this case (when a compensation circuit based on the Smith method is configured in an actual system), one calibration is required.

Therefore, the control circuit 3 performs calibration by adjusting the span and zero balance of each feedback signal 9. Note that this confirmation is performed by measurement at the output terminal of each operational amplifier of the control circuit 3, and the calibration procedure will be described below.

1. Perform zero point adjustment for each component.

2) Drive valve 1 when the maximum electrical operation signal (20mA) is input.
Adjust the span and gain on the input side so that the displacement is 8 countries.

3) Cut off the input electrical operation signal 4 to the control circuit 3.

4) Supply air pressure directly to the control valve 12.

5) Supply air until the zero point is adjusted.

6) TPFB by span D, C, +1.
Adjust to 0v.

7) Supply air until the displacement of the drive valve 1 reaches 8011.

8) By spanning TPFB, C, +5. Adjust to Ov.

9) Repeat steps 4) to 8) until the displacement of drive valve 1 is 0 to 81.
At m, adjust the span and zero balance on the feedback side so that the feedback signal 9 rises to +1 to 5v.

10) Perform the above procedure for three diaphragm-driven valves with electro-pneumatic converters.

After completing the calibration of each component, the following experiment was conducted to confirm that hunting due to the length of the pipe line was removed for the circuit shown in FIG.

The drive valve was operated with the compensator not operating at the pipe length Q=40rn, and the change in dynamic behavior of the hunting state was measured when the compensator was operated at an arbitrary time t=tc. Note that Table 1 shows the conditions of each diaphragm-driven valve used in this experiment.

In the table, the pipe lengths of Nα, 2, and N(1, 3) are determined by adjusting the air pipe resistance with a needle valve, taking into consideration the differences in the characteristics of the system used for the compensation circuit. change the value.

Table 1 When the input electrical operation signal 4 to the control circuit 3 is 20 mA, FIG. 9 shows the dynamic behavior of the driven valve 1 when the compensator is operated at an arbitrary time t=tc.

From FIG. 9, before operating the compensation circuit H' (t≦tc
), the displacement of the drive valve 1 is vertically asymmetric, and the period is 2.0.
4s, the amplitude is 2.89no, and the center position is yo=
An oscillation phenomenon of 8.87m+ is occurring. Therefore, when the compensator based on the Smith method is operated at t''tc, the displacement of the drive valve 1 is stable at 4.25 after the compensation circuit is operated (±1 of the steady value).
．． (within 0%), indicating that hunting has been removed.

In addition, the input electrical operation signal 4 to the control circuit 3 is set to 20III.
If it is reduced to A, it will take a little longer to stabilize after the compensation circuit H' is activated, and a steady-state deviation will remain.
It was confirmed that the displacement of the driven valve 1 was stabilized and the dynamic characteristics were improved.

FIG. 10 is a wiring diagram of the controller 3 used in the present invention,
It was constructed using electronic circuits based on data obtained from various tests using the control system shown in FIG. 8, and a compensation circuit H' based on the Smith method was added to the conventional control circuit.
It incorporates.

In FIG. 10, ■ is an input signal amplification section, J is an amplification section for the feedback signal from the potentiometer 1b, and H is an amplification section for the feedback signal from the potentiometer 1b.
' is a compensation circuit based on the Smith method, H forming a compensation circuit element
a and Hb are dead time adjustment circuits, Hc is a subtraction circuit that calculates the difference between the outputs of the two adjustment circuits Ha and Hb, K is P,
1. D is a control section, and P is an output amplification section.

Both dead time adjustment circuits Ha forming the compensation circuit H',
)(l) is subtracted by a subtraction circuit Hc, and then added to the feedback signal 9 by an arithmetic circuit M, and furthermore, the output is added to an input operation signal 4 by an arithmetic circuit N to obtain yPID.
is input to the control unit.

The output from the PID control unit K is output to the outside via the output transformer ○, and the piezoelectric element flapper 2 of the electro-pneumatic converter 2
input to a.

In addition, the output side to the PID control section is connected to both adjustment circuits Ha and Hb.
is connected to the input side of the

The operating characteristics of the automatic control valve with electropneumatic positioner using the compensation circuit H' shown in FIG. 10 are approximately the same as those shown in FIG.
By appropriately adjusting the compensation circuit H', controller output gain, input span, etc.

It has been confirmed that even if the pneumatic pipe line 8 extends over a long distance of 30 m or more, it is possible to perform extremely stable opening/closing control of the drive valve 1 without causing any oscillation phenomenon.

(Effects of the Invention) In the present invention, since the control circuit 3 is provided with the compensation circuit H' based on the Smith method, the transfer function of the controller 3 becomes independent of dead time. As a result, the stability of the control system is improved, and even if the pneumatic pipe line 8 becomes quite long, the opening/closing control of the driven valve does not become vibratory, and extremely stable valve opening/closing control can be performed. .

Further, the compensation circuit H' includes two variable dead time adjustment circuits Ha having the same type of transfer function as the controlled object.

By combining Hb and a subtraction circuit Hc for the outputs of both regulators, it can be extremely easily formed as an electronic circuit, and is excellent in preventing opening/closing vibration of an automatic control valve equipped with e l = positioner. It has many practical benefits.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a control system for an automatic control valve equipped with an electropneumatic positioner, which is an embodiment of the present invention. FIG. 2 is an explanatory diagram of the saturated nonlinear element in FIG. 1. FIG. 3 is a block diagram showing a control system for an automatic control valve provided with a compensation circuit according to the present invention. FIG. 4 shows the transient characteristics of the driven valve obtained from the simulation test based on FIG. FIG. 5 shows the transient characteristics of the drive valve in the response experiment apparatus of FIG. 13. FIG. 6 shows the oscillation characteristics of the driven valve in the response experiment apparatus of FIG. 13, and FIG. 7 shows the saturation characteristics of the input voltage and output voltage of the controller. FIG. 8 is a model configuration diagram of a control system including a compensation circuit. FIG. 9 shows the results of an experiment to stabilize the displacement of the drive valve 1 in the control system model of FIG. 8. FIG. 10 is a circuit diagram of a controller incorporating a compensation circuit used to implement the present invention. FIG. 11 is an explanatory diagram of the operating principle of a conventional automatic control valve with an electropneumatic positioner. FIG. 12 is a schematic explanatory diagram of an electro-pneumatic converter. FIG. 13 is an explanatory diagram of a response experiment apparatus for an automatic control valve with an electropneumatic positioner. FIG. 14 is a block diagram of a control system for an automatic control valve with an electropneumatic positioner, and FIG. 15 is an implementation diagram of the feedback circuit of equation (4). FIG. 16 is a configuration diagram of a control system in which the compensation circuit based on the control system Hesmith method of FIG. 14 is applied. FIG. 17 shows a converted form of the Smith method control system. 1 Diaphragm driven valve 1a Diaphragm drive unit 1b Linear potentiometer 1c Stem 2 Electropneumatic converter 2a Piezoelectric element flapper 2b Nozzle 3 Controller 4 Human electric operation signal 5 Control signal 6 Pilot valve 6a Pressure chamber 6b Back pressure chamber 6c Spool 6d Port 7 Active air 8 Pneumatic pipe 9 Displacement signal (feedback signal) 8 Controller transfer function B Electro-pneumatic converter transfer function C Delay dead time (L) transfer function D Drive valve transfer function E Saturated nonlinear Element F Transfer function G of the linear potentiometer Transfer function H of the controlled object Transfer function H' of the compensation circuit Compensation circuit Ha, Hb variable dead time adjustment circuit Hc Subtraction circuit Input signal amplification section J Feedback signal amplification section K PID control section L Dead time M-N operational amplifier circuit O Hatake transformer Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 5 30 60 (ff) Figure 6 Figure 7 l; (v) Fig. 8 Fig. 9 Fig. 30 60 (of course Fig. 11 Fig. 12 Fig. 13

Claims (2)

[Claims] (1) The control signal (5) from the controller (3) activates the electro-pneumatic converter (2) equipped with a flapper (2a) made of piezoelectric element, and connects the driven valve (1) through the pneumatic pipe (8). In addition to supplying driving air pressure to the linear potentiometer (
By feeding back a displacement signal (9) proportional to the displacement of the stem (1c) of the drive valve (1) detected by the controller (1b) to the controller (3), the displacement of the stem (1c) is controlled by the controller (3). ), in an automatic control valve equipped with an electropneumatic positioner which is held in a position corresponding to an input electrical operation signal (4) to the controller (3), there is a In parallel, transfer functions Gf(s)・Gp(s)・(1-e^-^s^L) [However, Gf(s) is the transfer function of the linear potentiometer (1b), and Gp(s) is the dead time. The transfer function of the controlled object without (8) A method for preventing oscillation of an automatic control valve equipped with an electro-pneumatic positioner, characterized by preventing opening/closing vibration of the drive valve (1) when lengthening the positioner. (2) The compensation circuit (H') is replaced by two variable delay time (L) adjustment circuits (Ha), (Hb) which have the same type of transfer function as the controlled object and whose input sides are connected in parallel. and a subtraction circuit (Hc) for the outputs of both the adjustment circuits (Ha) and (Hb).
) A method for preventing oscillation of an automatic control valve equipped with an electropneumatic positioner according to claim (1).