JPH036701A - Controller - Google Patents

Abstract

PURPOSE:To improve the control response and to ensure the stable and safe control for a controller applied to an instrumentation system by changing forcibly the integration control arithmetic output in accordance with the degree where a control signal exceeds the signal control value. CONSTITUTION:A difference signal MVn - 1 between the preceding input signal and the preceding output signal applied to a changing rate control means 21 is obtained by a difference signal acquiring means 22 and sent to an addition means 23. At the same time, an addition signal (P + D)n of the proportion control arithmetic output Pn and the differentiation control arithmetic output Dn is supplied to a changing component arithmetic means 20. The changing component DELTA(P + D)n between the preceding value and the current value of the means 20 is obtained and added to the signal DELTAMVn - 1. This addition signal is compared with the changing rate limit value which is previously judged by a changing rate variance judging means 24. If it is judged that the addition signal is kept out of the limit value of the changing rate, a switching means 16 is set in an inconductive state to stop the integration control arithmetic output and to hold the preceding integration control component. Thus store effect of the integration control component is eliminated and the overshoot can be prevented.

Description

Detailed Description of the Invention [Objective of the Invention] (Industrial Application Field) The present invention relates to a regulating device used in various process instrumentation control systems, and particularly relates to a regulating device that is used in various process instrumentation control systems, and particularly relates to a regulating device that is used in various process instrumentation control systems, and particularly relates to a regulating device that is used in various process instrumentation control systems.
The present invention relates to an adjustment device that improves control responsiveness when an adjustment signal obtained by an adjustment calculation (I: integral, D: differential) exceeds a predetermined signal limit value.

(Prior Art) This type of PID control device is used in all industrial fields, and among them, it is indispensable for plant operation control.

Conventionally, there have been two types of PID calculation methods, an analog calculation method and a digital calculation method, of which the analog calculation method was often used until recently, but the digital calculation method has now become mainstream.

By the way, the PID algorithm in the analog calculation method performs calculations according to the following basic formula.

Mv-Kp (e+ (1/T+), fedt+T
o (de/dt)+MV.

...(1) However, in the above formula, MV is the manipulated variable and e is the deviation.

K is the proportional gain, TI is the integral time + TD is the differential time, and MVo is the initial value of the manipulated variable.

Therefore, when executing a digital calculation method using the basic formula of the PID algorithm in equation (1), the sampling period τ is determined in advance, and the necessary data is acquired for each sampling period τ and calculations are performed. . For example, if the current sampling point is nτ (n is an integer) and the previous sampling point is (n - 1)τ, then the deviation of the current sampling point obtained from the process control system is e
The deviation of n s previous sampling times can be expressed as e n-1.

Moreover, two types of calculation methods are considered for this digital calculation method, one of which is a position type calculation method.
The other one is a velocity type calculation method.

The former position type calculation method is a method that directly calculates the entire manipulated variable M V n at each sampling time.
The latter speed type calculation method calculates only the changed manipulated variable △MVn at the current sampling time, and calculates the previous manipulated variable MVn-
In this method, the current operation m MV n is obtained by adding the currently obtained change operation amount ΔM V n to 1. Therefore, the velocity type calculation method is MVn - MVn-1 + △M
It is expressed as Vn.

Therefore, when executing the position type calculation method and the velocity type calculation method according to the basic formula of the PID algorithm of formula (1), in the position type calculation method, + (To/r) (en
-en-1)) + MVo -(2) The calculation is performed using the PID algorithm, while in the velocity type calculation method, ΔMVn -Kp ((en -en-1) + (r
/T+)en The calculation is performed based on the PID algorithm.

As far as the PID algorithms of equations (2) and (3) are considered, the velocity-based PID algorithm of equation (3) is superior to the position-based algorithm of equation (2) in various aspects.

The reason for this is that Σ is removed from the integral term, which simplifies calculations, and when switching the digital controller itself from manual to automatic, the current manipulated variable is expressed as M V n in equation (3).
-1 to perform automatic control and add the change operation amount △M V n to M V n-1 from the next sampling point, so it is easy to perform balanceless/bumpless switching between manual and automatic switching. In addition, if a limit is given in advance to the previous actual operation j1MVn-1, the integral term will not be accumulated to an extreme degree, and so-called reset wind-up due to the integral term can be easily prevented, and furthermore, it is possible to easily prevent the so-called reset windup due to the integral term. There is no need to limit the output change or modify the gain because only the
It also has the advantage of being able to easily perform complex arithmetic processing with other signals, and from these points of view, DDC (Direct
t D1digital Control! ), velocity type calculation methods are widely used.

However, regardless of whether an analog calculation method, a position-based digital calculation method, or a velocity-based digital calculation method is adopted, in order to create a control system that can actually be used industrially, the operating end of the controlled object, the controlled quantity (steam , fuel, water, air, etc.)
A signal limiting means is provided so as not to cause sudden changes to the control system and downstream processes, and an operation signal obtained from the adjustment calculation output through the signal limiting means is applied to the controlled object. This signal limiting means is indispensable to perform safe and stable control in addition to the constraints, characteristics, and control needs of the controlled object.

Therefore, as mentioned above, if there is no restriction means in various calculation methods, there will be no problem and the safety and stability will be lacking, but the inventor has repeatedly conducted simulation experiments. It has been found that even if restrictions are placed on the adjustment signal in the process, it is not uniformly safe and stable, and control responsiveness significantly decreases depending on the state of the restriction.

Next, a conventional composite digital PID calculation system adjustment device that combines a velocity type calculation system and a position type calculation system will be described with reference to FIG.

That is, in this adjustment device, after the subtraction means 1 subtracts the current process variable value PVn from the current target value SVn to obtain the deviation en, this deviation en is calculated by the position type proportional adjustment calculation means 2. It is introduced into the speed-type integral adjustment calculation means 3 and the position-type differential adjustment calculation means 4, respectively. This position type proportional adjustment calculation means 2 performs adjustment calculation of the proportional term of the above equation (2), that is, p-Hl, and converts the obtained proportional adjustment calculation component into tx
Introduced into J Calculation Means 5. In addition, the velocity type integral adjustment calculation means 3 calculates the integral term of the above equation (3), that is, △In-KIJ・
After performing the adjustment calculation (r/T+) en to obtain the integral adjustment component △In, convert this integral adjustment component △In into the velocity form -
The signal is introduced into the position shadow signal converting means 6, and the calculation In-1n-1+△In is performed to convert it into a position shadow signal In, which is then sent to the adding means 5.
to be introduced. Furthermore, the position type differential adjustment calculating means 4 calculates the differential term of the equation (2), that is, Dn −Kp ·(Tp
/r) (en - en - 1) to obtain the differential adjustment component Dn, which is similarly introduced into the adding means 5.

After obtaining the proportional, integral, and differential adjustment components as described above, the addition means 5 supplies these adjustment components to the change rate limiting means 7. This change rate limiting means 7 controls the adjustment signal M
Operation signal M in which V n is adapted to the constraint conditions of the controlled object 8
After setting V n, apply it to the controlled object 8 and en-Q,
In other words, control is performed such that the voltage becomes 5Vn-PVn.

(Problem to be Solved by the Invention) In the conventional adjustment device, when the target value SVn or the process variable value due to disturbance suddenly changes, the adjustment signal MVn theoretically changes greatly and is applied to the controlled object 8. ,
As a result, each component that makes up the control system, such as the operating end and piping system, is damaged.

This will lead to a shortened lifespan and a decline in product quality, but the third
When the rate of change limiting means 7 is provided as shown in the figure, the rate of change of the adjustment signal sent from the adjustment calculation means is limited under the constraint conditions of the controlled object 8, which causes various problems such as those mentioned above. can be prevented.

However, when the present inventor actually installed the rate-of-change limiting means 7 and conducted simulation experiments, it was found that the control response was significantly affected in addition to the conditions shown in FIG. 4, depending on the rate-of-change limiting state. I found out. In other words, (
A) is the PV response when the target value SVn is changed by 20% with the change rate limit value ΔMVH - 10%, (A)'
shows the MV response under the same conditions, and (b) in the same figure shows the target value SV with the rate of change limit value ΔM V H = 396.
PV response when n is changed by 20%, and (b)' shows MV response under the same conditions. Therefore, as is clear from this response characteristic, even when the PID parameters are the same, the control response changes greatly depending on the rate of change limit value, and contrary to general speculation, the overshoot increases. occurs.

Therefore, in the conventional device as described above, (1) the control response changes depending on the magnitude of the change rate limit value with respect to the operation signal, and particularly when the limit value is made small, the overshoot becomes large.

This poses a problem that adversely affects plant safety and product quality.

■Therefore, it is necessary to adjust the optimal PID parameters according to the magnitude of the rate of change limit value for the operation signal.
General CHR (K, L, C
Hicn, J.A.) Irones, J.B.

Since PID parameter adjustment formulas such as the Re5vick method cannot be applied, the PID parameter adjustment work becomes extremely troublesome.

The present invention has been made in view of the above circumstances, and by forcibly changing the integral adjustment calculation output according to the extent to which the adjustment signal exceeds the signal limit value, control responsiveness can be improved, and it can be applied to instrumentation systems. The purpose of the present invention is to provide a swing adjustment device that can be controlled stably and safely.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above problems, the invention according to claim 1 uses a deviation between a process variable value from a controlled object and a target value to - An adjustment device that performs integral or proportional/integral/derivative adjustment calculations to obtain an adjustment signal, and applies this adjustment signal to the control object by applying an operation signal obtained through the signal restriction means to the signal restriction means. A difference signal acquisition means for obtaining a difference signal between a previous input value and a previous output value, and an adjustment change calculation means for calculating a change in the proportional or proportional/derivative adjustment calculation output from the previous value to the current value, further comprising: The outputs obtained by the difference signal acquisition means and the change calculation means are added by the addition means, and the signal deviation judgment means determines whether or not this added signal exceeds a predetermined limit value by the signal restriction means. and an integral output hold setting means for stopping the integral adjustment calculation output and setting the integral adjustment component to a hold state when the signal deviation determination means determines that the signal exceeds the signal limit value. It is.

Next, the invention recited in claim 2 is configured to further input and add a change in the integral adjustment calculation output from the previous value to the current value to the addition means of the invention recited in claim 1. be.

(Operation) Therefore, by taking the above-described measures, the present invention calculates the deviation between the process variable value and the target value using the deviation calculation means, and then adjusts this deviation by proportional/integral or proportional/integral/derivative adjustment. The calculation is performed to obtain an adjustment signal, and this adjustment signal is restricted by the signal restriction means to control the controlled object as an operation signal. At this time, the previous input signal and the previous output signal to the signal restriction means are taken in. Then, a difference signal between the two signals is obtained by the difference signal obtaining means and introduced into the addition means. On the other hand, the proportional/differential adjustment component, which is the addition signal of the proportional adjustment calculation output and the differential adjustment calculation output, is introduced into the change calculation means, and here the change from the previous value to the current value of the proportional/differential adjustment component is calculated. and then added to the difference signal.

The sum signal of the difference signal and the change signal obtained in this way is compared with a rate of change limit value predetermined by the rate of change limiter in a change rate deviation determining means, and the sum signal deviates from the rate of change limit value. determine whether or not. Here, if it is determined that there is no deviation, the integral adjustment calculation output is used as the operation signal, and if it is determined that there is a deviation, the integral adjustment calculation output is forcibly stopped, and the previous integral adjustment component is hold. In this way, by selectively using normal integration and stopping integral operation depending on the degree to which the adjustment signal exceeds the rate of change limit value, PI control can be achieved while inheriting the essence of PID control.
Overshoot can be eliminated by removing the cumulative effect of the integral adjustment component without changing the D parameter.

(Example) Hereinafter, an example of the present invention will be described with reference to FIG. In the figure, 11 is the target value SVn and the controlled object 12.
A deviation calculating means for calculating the deviation eQ from the process variable value PVn from were introduced, respectively.
These calculating means 13, 14, and 15 each calculate the deviation eQ.
In response to this, integral adjustment calculation operation, proportional adjustment calculation operation, and differential adjustment calculation operation are performed. On the output side of the speed type adjustment calculation means 13, a speed type/position shadow signal conversion means 17 is provided via a switching means 16 which is normally in a conductive state. The signal is converted into a form integral signal and introduced into the subsequent adding means 18.

Reference numeral 19 denotes an addition means for adding and combining the output of the position type proportional adjustment calculation means 14 and the output of the position type differential adjustment calculation means 15, and this addition signal is introduced into the addition means 18 and the change amount calculation means 20. There is.

This addition means 18 adds and synthesizes the proportional, integral, and differential adjustment components to obtain an adjustment signal, and transfers this adjustment signal to the change rate limiting means 2.
1, and here it is limited to conform to the constraint conditions of the controlled object 12 and applied to the controlled object 12 as an operation signal, and en
-0, that is, SVn -PVn.

On the other hand, the change amount calculation means 20 includes each calculation means 14.15.
It has a function to subtract the previous value of the proportional/differential adjustment component from the current value of the proportional/differential adjustment component obtained from , to obtain the change in the proportional/differential adjustment component.

Reference numeral 22 denotes a difference signal acquisition means, which extracts a difference signal between the previous input signal and the previous output signal of the change rate limiting means 21 and introduces it into the addition means 23.

The addition means 23 adds and synthesizes the difference signal obtained by the difference signal acquisition means 22 and the change in the proportional/differential adjustment component obtained by the change calculation means 20, and then sends the resultant signal to the change rate deviation judgment means 24. This rate of change deviation determination means 24 includes the addition means 2
The change rate limiting means 21 determines whether the addition signal from 3 exceeds a predetermined change rate limit value, and when it is determined that the addition signal does not exceed the change rate limit value, the switching means 16 is made conductive. The output of the speed-type integral adjustment calculating means 13 is directly led to the signal converting means 17 to perform so-called normal integration, and on the other hand, when it is determined that the added signal by the adding means 23 exceeds a predetermined rate of change limit value, the switching is performed. The configuration is such that the means 16 is set to a non-conducting state, that is, the integral adjustment calculation operation is stopped, and the signal converting means 17 sets the previous integral adjustment component to a hold state.

Next, the operation of the apparatus configured as above will be explained.

The deviation calculation means 11 calculates the target value SVn and the controlled object 12.
Taking in the process variable value PVn of (SVn - PV
After obtaining the deviation signal en by performing the calculations shown in FIG. here,
When the position type proportional adjustment calculation means 14 receives the deviation signal en, it performs a proportional adjustment calculation operation of Pn - kp IIen based on the proportional term of equation (2) to obtain a proportional adjustment component Pn,
This proportional adjustment component Pn is introduced into the adding means 19. In addition, the position type differential adjustment calculation means 15 also outputs the deviation signal en.
When received, the calculation of the differential term in equation (2) above, that is, Dn − kp ・(To /r) ・
After performing the differential adjustment calculation operation (en - en - 1) to obtain the differential adjustment component Dn, the differential adjustment component Dn is led to the addition means 19, where the differential adjustment component Dn and the previous proportional adjustment component Pn are added and synthesized, The obtained addition signal (P+D)n is introduced into the addition means 18 and the change calculation means 20.

On the other hand, the speed-type integral adjustment calculation means 13 calculates the integral term of the above equation (3) based on the deviation signal en, that is, performs the integral adjustment calculation operation ΔIn −kp · Cr/TI) ·en to obtain the integral adjustment component ΔIr+ is converted into a speed shadow signal ΔIn' through the switching means 16 which is normally in a conductive state, and then converted into a speed shadow signal ΔIn'.
7, where it is converted into a position type signal In based on the following formula.

In-In-1+Δ In' Then, the position signal In converted by the signal converting means 17
is sent to the addition means 18 as an integral adjustment component. The adding means 18 adds and synthesizes the integral adjustment component In and the proportional/differential adjustment component (P+D)n, and then outputs the resultant signal to the controlled object 12 as an operation signal M V n via the change rate limiting means 21.
is applied to control the deviation en-0, that is, 5Vn-PVn.

When controlling the controlled object 12, the difference signal acquisition means 22 uses the previous input signal M V n-1 and the previous output signal M V n
-1 and performs the calculation ΔMVn-1=MVn-1-MVn() to obtain the difference signal ΔMVn-1 and send it to the adding means 23. Therefore, at this time, the current proportional/differential When the adjustment component (P+D)n is introduced into the change calculation means 20, the change calculation means 20 calculates the current proportional/differential adjustment component (P+D)n and the previous proportional/differential adjustment component (
Using P+D)n-1, △(P+D) n -(P+D) n -(P+D)
The change Δ in the proportional/differential adjustment component is determined by the calculation formula n-1.
(P+D)n is obtained, and this change Δ(p+D)n is introduced into the adding means 23. As a result, the addition means 23 outputs an addition signal of 6M V n-1+Δ(p+D) n , which is introduced into the change rate deviation determination means 24 . Therefore, in this change rate deviation determination means 24, the addition signal ΔMVn-1+Δ(P+
D) n is compared with the rate of change limit value ΔMVH set in advance by the rate of change limiter 21, and the addition signal ΔM V n-1+Δ
It is determined whether or not (P+D)n deviates from the rate of change limit value ΔMV, and when it is determined that it does not deviate, the switching means 16 is made conductive, and the speed-type integral adjustment calculation means 13 obtains the The integral adjustment component ΔIn is guided through the switching means 16 as a signal Δln/ to the signal converting means 17 to perform normal integration.

On the other hand, when it is determined that the addition signal ΔM V n-1+Δ(p+D)n deviates from the rate of change limit value ΔMVH, the switching means 16 is set to a non-conducting state, the integrating operation is stopped, and the signal converting means 17 The previous integral adjustment component is held. In other words, when 1ΔMVn-1+Δ(P+D)nl≦ΔMVH, the rate-of-change deviation determination means 24 determines that the predetermined rate-of-change limit value ΔMVH is not exceeded, and executes normal integration, so that ΔMVH<lΔM When V n-1+Δ(P+D) n, it is determined that the predetermined change rate limit value ΔMVH is exceeded, the integral operation is stopped, and the integral adjustment component is set to a hold state.

Therefore, according to the configuration of the embodiment as described above, the change in the proportional, integral, and differential adjustment signals can be adjusted depending on the magnitude relationship with the change rate limit value predetermined based on process constraints, control needs, etc. Forcibly intervenes in the integral operation during the adjustment operation and selects and executes either stopping the integration or normal integration, so it is possible to settle without overshooting the step response of the target value, as shown in Fig. 4, for example. . That is, in the conventional device, as is clear from FIGS. 4(a) and 4(b), as the rate of change limit value is made smaller, the overshoot in the control response gradually increases. For example, when the rate of change limit value is 10%, the step response of the target value as shown in the diagram (a) is shown, and the manipulated variable at that time is (a)', whereas when the rate of change limit value is 3%, Sometimes, a manipulated variable such as the step response 1 of the target value shown in the figure (b)' is obtained. This means that when the target value is changed, the operation signal M V n would originally want to change greatly, but because it is limited by the rate of change, it gradually increases. During this period, the deviation en obtained by the deviation calculation means 1 increases, but this is because the speed type integral adjustment calculation means 3 performs integral adjustment calculation and accumulates the deviation en while integrating it. Therefore, as the deviation en decreases over time, the proportional and differential adjustment calculation components become zero, but only the accumulated amount due to the integral adjustment calculation remains, which greatly exceeds the step response of the target value. You end up letting him shoot.

Furthermore, when considering the essence of PID control based on the above equation (1), the following table can be compiled.

In other words, from the perspective of inheriting the essence of PID control, when the change in the adjustment signal exceeds a predetermined rate of change limit value, it is possible to truncate the signal in integral operation, but in the case of proportional operation, the signal If you truncate the signal, it will not become zero when the deviation signal 8 is zero, and in the case of differential operation, if you truncate the signal, it will not become zero when there is no change.

Therefore, when considered from the above two perspectives, by selecting normal integration or stopping the integral operation depending on the degree to which the change in the adjustment signal exceeds a predetermined rate of change limit value, the essence of PID control can be reliably inherited. In addition, overshoot of the control response due to the rate of change limit value can be avoided without changing the PID parameter.

Next, an embodiment of the invention according to claim 2 will be described with reference to FIG. 2. That is, this device has almost the same configuration as that in FIG. 1, and the particular difference is that the integral adjustment component ΔIn of the speed-type integral adjustment calculation means 13, that is, ΔIn −kp # (r/Tl ) · en By applying this to the addition means 23, a signal of (6M V n-1+Δ(P+D)n+ΔInl) is input to the rate of change deviation determination means 24, thereby increasing the determination accuracy.

That is, in the change rate deviation determination means 24 of this device, ΔMVn-1+Δ(P+D)n +ΔIn≦ΔMV
When H, an on control signal is sent to set the switching means 16 to a conductive state, and a normal integral adjustment component is sent to the signal conversion means 17, while 6M V n-1+△(P+D)n +Δrn〉△ MV
When H, an off control signal is sent to set the switching means 16 to a non-conducting state to stop the integral adjustment operation, and the signal conversion means 1
7, the integral adjustment component is held. Generally, △In is smaller than Δ(P+D)n, so
Although ΔIn may be regarded as approximately zero and omitted as in the invention described in claim 1, if the processing is performed as in the present apparatus, the determination of the integral operation will be more accurate.

In the above embodiment, the proportional/integral/differential: A section calculation operation was described, but it may also be a proportional/integral adjustment calculation operation, for example, and in this case, one of the adding means is unnecessary, and the proportional adjustment Proportional adjustment component Pn of calculation means 14
is directly introduced into the addition means 18 and the change calculation means 20.

In addition, the present invention can be implemented with various modifications without departing from the gist thereof.

[Effects of the Invention] As explained above, according to the present invention, the change in the proportional/integral adjustment or the proportional/integral/derivative adjustment signal is a change set in advance in consideration of process constraints, control needs, etc. Forcibly intervenes in the integral adjustment component in the adjustment signal depending on the magnitude relationship with the rate limit value, and selects and executes either "stop integral adjustment component" or "output normal integral adjustment component" As described above, control can be performed while taking advantage of the essence of PID control, and as shown in (c) of Fig. 4, when the rate of change limit value is 3%, the rate increases while showing a step response similar to that of the conventional device. When the absolute value of the adjustment signal exceeds the rate-of-change limit value, the switching means 16 is brought into a non-conducting state so that it temporarily follows the proportional and differential adjustment components over time, but there is no integral effect due to the integral operation. Therefore, there is no overshooting of the target value, overshooting of the control response can be prevented, and stable and safe control can be achieved.

In addition, there is no need to re-tune the PID parameters in response to changes in the rate of change limit value, and the CHR method, which is a general PID parameter adjustment formula, can be used as is, making it easy to change the rate of change limit value. Therefore, it is possible to simplify the PID parameter tuning work.

Therefore, according to the above-mentioned adjustment device, it is possible to make effective use of the PID adjustment function, which is most commonly used in plant control systems, and to make effective use of the rate-of-change limiting function, while making plant operation more flexible in the future. , optimal control, super automation, etc., and can make a significant contribution to a wide range of industrial fields.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing one embodiment of the adjustment device according to the present invention, Fig. 2 is a block diagram showing an embodiment of another invention, Fig. 3 is a block diagram of a conventional device, and Fig. 4 is a block diagram of a conventional device. FIG. 3 is a diagram illustrating a comparison of control responses between the device and the device of the present invention. 11... Deviation calculation means, 12... Controlled object, 13.
...Velocity type integral adjustment calculation means, 14...Position type proportional adjustment calculation means, 15...Position type differential adjustment calculation means, 16
...Switching means, 17...Signal conversion means, 1
8゜19...addition means, 20...change calculation means,
21 change rate limiting means, 22... difference signal acquisition means, 23
... Addition means, 24... Change rate deviation judgment means.

Claims (2)

[Claims] (1) Obtain an adjustment signal by performing proportional/integral or proportional/integral/derivative adjustment calculations using the deviation between the process variable value from the controlled object and the target value, and use this adjustment signal as an operation signal obtained through the signal limiting means. The adjustment device controls the control object by applying it to the controlled object, comprising: a difference signal acquisition means for obtaining a signal of the difference between the previous input value and the previous output value to the signal limiting means; a change calculation means for calculating the change from the previous value to the current value; an addition means for adding the change signal obtained by the change calculation means to the difference signal obtained by the difference signal acquisition means; and the addition means. a signal deviation determining means for determining whether or not the added signal obtained by the above signal exceeds a predetermined limit value by the signal limiting means; An adjustment device comprising: an integral output hold setting means for stopping an integral adjustment calculation output and setting an integral adjustment component to a hold state. (2) The adjusting device according to claim 1, wherein the adding means adds a change in the integral adjustment calculation output from the previous value to the current value.