JP2645112B2 - Gain adaptive control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Purpose of the Invention] (Field of Industrial Application) The present invention relates to a gain adaptive control device using a process instrumentation control system or the like, and in particular, responds to disturbance changes, process characteristic changes, and the like. The present invention relates to a gain adaptive control device that stably controls an object to be controlled.

(Prior art) In recent years, fluctuations in production volume, load, and the like are accompanied by various changes in demand. One of the major themes is how to suppress the effects of these fluctuations at high speed, high accuracy, and stably. I have.
In particular, in the future, it is thought that there will be a strong demand for the emergence of full-scale flexible automation that can flexibly follow changes in product type, quality, etc., in addition to production volume.

Therefore, in order to respond to the above demands, in addition to a feedback (hereinafter abbreviated as FB) control system, a feed system that detects a disturbance such as a production amount and a load and gives a signal for canceling the influence of the disturbance in advance. It has become important to effectively utilize a so-called FF / FB control system that combines a forward (hereinafter abbreviated as FF) control system.

However, even in this FF / FB control system, the FF gain, which indicates how much the manipulated variable is changed with respect to the change in disturbance, is not always constant. Then, the FF gain changes accordingly. Therefore, in order to always execute the FF / FB control system under optimal conditions, the FF gain, FB
It is necessary to automatically correct the gain so that it is always optimal.

By the way, in general, in an FB control system, P (proportional) +
Since I (integral) operation control is mainly performed and D (differential) operation control is added as an auxiliary function, hereinafter, in the case of FB control, PI
Description will be given as arithmetic control.

Now, the basic PI calculation formula of the FB control in the analog control method is represented by MV = kpke + (1 / T _I ) ∫edt｝ + MV ₀ (1) Here, MV is the manipulated variable, e is the deviation, kp is the proportional gain, T _I is the integration time, and MV ₀ is the initial value of the manipulated variable.

By the way, in the case of executing the digital control method using the PI operation expression represented by the above expression (1), a sampling period τ is determined in advance, and each data is taken in every sampling period τ to perform the operation. Become. Therefore, if the current sampling time is nτ (n is an integer) and the immediately preceding sampling time is (n-1) τ, the deviation of the current sampling time obtained from the control system is en, and the deviation of the previous sampling time is en Can be represented by en-1.

The digital control method has two operation methods, one of which is a position-type operation method for obtaining the total manipulated variable MVn at each sampling time, and the other is the change amount at the current sampling time. ΔMVn alone is obtained, and this change amount is added to the previous operation amount MV n−1 to obtain the current operation amount MV.
This is a speed type calculation method with n.

Here, the former position type calculation method is as follows. On the other hand, the latter velocity type calculation method is as follows: ΔMVn = kp ｛(en−en−1) + (τ / T _I ) · en｝ (3) MVn = MVn−1 + △ MVn .. (3 '), each of which has its own characteristics.

Furthermore, in the process control system, if the FF control system is always in the optimum state, the FB control output = 0, that is, the dynamic component is rather harmful. Operation output (MVn) ... It is only necessary to correct the gain so as to satisfy the relationship of (4).

Therefore, when obtaining the gain correction coefficient from the above relationship, as shown in FIG. 3, the FF control output signal FF n-1 is changed to the operation output signal MV n-1 (= (n- 1) Assuming that a gain optimum correction coefficient for making equal to the optimum correction output signal Xn-1) at the time is kn-1, MVn-1 = kn-1.FF n-1 kn-1 = (MVn- 1) / (FF n-1) (5)

Next, when the FF control output signal FFn-1 changes to FFn, the FF control optimum output signal Xn is given by: Xn = kn-1 (optimum correction coefficient at the time point n-1) FFn (FF control output at the time point n) Signal) = (MV n−1 / FF n−1) · FFn (6) Since this equation (6) is a position-type operation method, if it is converted into a speed-type operation method, Xn = Xn-1 + △ Xn △ Xn = Xn-Xn-1. Here, X n shown in the above equation (6) and FIG.
-1 = MV n-1 to △ Xn = (MV n−1 / FF n−1) · FFn−MV n−1 = (MV n−1 / FF n−1) · (FFn−FF n−1) = (MV n−1 / FF n−1) · △ FFn (7) is obtained.

By the way, this type of conventional control device is configured as shown in FIG. 4 based on the above various conditions and arithmetic expressions. Process amount detector 2 on the output side of control target 1
The process quantity PVn detected by the process quantity detector 2 and the target quantity SVn from the target quantity setting unit 3 are introduced into the deviation calculating means 4, where the calculation of en = SVn-PVn is performed to calculate the deviation. The en is obtained and sent to the speed type PI adjustment calculating means 5. Here, the speed-type PI adjustment calculating means 5 calculates according to the equation (3) to obtain the current manipulated variable change ΔCn, and then introduces this ΔCn into the adding means 6. Then, the output △ MVn of the adding means 6 is guided to the speed-position-type signal converting means 7, where it is operated according to the equation (3 ') to be converted into the position-type operation signal MVn and applied to the control target 1. It constitutes a so-called FB control system, and further combines this FB control system with an FF control system.

In this FF control system, a disturbance signal Dn such as a production amount or a load is detected by a disturbance amount detector 10, and the disturbance signal Dn is input to a coefficient means 11 to multiply a coefficient to obtain a static FF control signal FFn. It is supplied to the shape-speed signal conversion means 12. This conversion means 12
Then, the speed type signal of the difference from the previous value and the current value △ FFn = FFn-
FF n−1 is obtained and multiplied by the speed type signal △ FF of the difference.
Introduce to 13.

On the other hand, the static FF control output signal is sent to the gain correction coefficient calculating means 14 together with the operation signal, where the gain correction coefficient kn-1 shown in the equation (7) = MV n−1 / FF n−1, that is, the timing To find the optimum value this time, find the previous gain correction coefficient Kn-1 and introduce it to the multiplying means 13 to multiply the previously obtained change 静 的 FFn of the static FF control output signal to obtain kn−
1 · △ FFn is obtained and further added and synthesized by the adding means 6 with the FB control output signal 加 算 Cn, so that the FB control system
The control target 1 is controlled by combining the controls.

(Problems to be Solved by the Invention) However, the above-described control device has the following various problems.

(A) One of them is that the operation signal MV includes a P component and an I component of the FB control system as shown in FIG. 5A, and among them, the P component becomes kp · en. Fluctuates drastically as shown in FIG. FIG. 2B shows the I component. Accordingly, when the frequency fluctuates as shown in FIG. 5 (a), the gain correction coefficient kn is correspondingly changed.
-1 greatly fluctuates, so that accurate and stable gain correction cannot be performed. In this respect, a great effect appears particularly when the target amount changes stepwise.

(B) Another is that when the PI is also a speed type operation, the operation signal MV gradually increases due to the integration operation and is caught by the upper / lower limit function added to the speed type-position type signal conversion means 7. In such a case, the operation signal is truncated, so that when the data is recovered again, the truncation reduces controllability. At this time, since the P component, which has particularly high variability, is discarded, the essence of the P control that the P component becomes zero when the deviation is zero is distorted, and the P component does not become zero, and the controllability and stability deteriorate. I do. For this reason, a method in which both P and I are speed-type calculations is not suitable for industrial use.

The present invention has been made in view of the above circumstances, and provides a gain adaptive control device that can always accurately and stably correct a gain, and that can enhance controllability and stability by utilizing characteristics of PI control. The purpose is to:

[Configuration of the Invention] (Means for Solving the Problems) In order to achieve the above object, according to the present invention, after a deviation is calculated by a deviation calculating means using a process amount and a target value, a PI is calculated based on the deviation. (P: proportional, I: integral) In a control device in which a feedforward control system is added to a feedback control system that obtains an operation amount to control a controlled object by performing an operation, a speed system integration operation is performed based on the deviation. Speed-type integral adjustment calculating means, a position-type proportional adjustment calculating means, which is provided separately from the speed-type nodal adjustment calculating means, and performs a position-type proportional calculation based on the deviation, and multiplies a disturbance amount by a feedforward coefficient. A first signal converting means for obtaining a static feedforward control signal and converting the control signal into a speed-type signal which is a difference; and outputting the output of the speed-type integral adjustment calculating means to a second signal converting means. Gain correction coefficient calculating means for obtaining a gain correction coefficient using the converted position type signal and the static feedforward control signal; and multiplying the speed type signal as the difference by the gain correction coefficient obtained by the calculation means. Compensating means for performing disturbance compensation by adding the obtained static characteristic compensation signal to the output of the speed-type integral adjustment calculating means; and converting the signal compensated by the compensating means to the second signal.
A gain adaptive control device comprising an operation amount obtaining unit that adds an output of the position type operation amount converted by the signal conversion unit and an output of the position type proportional adjustment operation unit to obtain an operation amount signal to a control target. is there.

Further, the present invention has a configuration in which a multiplying means is provided between the deviation calculating means and the speed-type integral adjustment calculating means, and the deviation is multiplied by a gain correction coefficient to perform FB control type gain correction.

(Operation) Accordingly, the present invention, by taking the above-described means, separates the PI operation into a speed-type integral (I) adjustment operation and a position-type proportional (P) adjustment operation and outputs the speed-type integral adjustment operation output. Is converted into a position-type signal and the position-type proportional (P) adjustment calculation output is added to generate an operation signal, so that control can be performed even if the operation signal recovers after exceeding the upper and lower limit values. The output of the speed-type integral adjustment operation means, that is, the average signal without fluctuation, is introduced into the gain correction coefficient operation means, and the FF gain is corrected using the gain correction coefficient obtained here. Thus, a stable and accurate gain correction can always be performed.

(Embodiment) Hereinafter, an embodiment of the present invention will be described with reference to FIG. This device has a block configuration for correcting only the gain of the FF control, and is used for controlling, for example, a flow rate, a pressure, a level, and the like. In the figure, 21 is a process amount detector, 22 is a target amount setting unit, and 23 is a process amount detector 21.
A deviation calculating means for calculating a deviation between a detection process amount and a target amount from the above. The obtained deviation is divided into two, one of which is provided to a speed-type integral adjustment calculating means 24 and the other is provided for a position-type proportional adjustment calculation. Introduced into means 25. The speed-type integral adjustment calculating means 24 performs a speed-type integral calculation on the deviation output from the deviation calculating means 23 based on the above equation (3). On the other hand, the position type proportional adjustment calculating means 25 performs a proportional calculation of multiplying the deviation by a proportional gain. 26 is an adding means for adding the output of the calculating means 24 and the FF control output, and this added output is sent to a speed-position-type signal converting means 27 for converting into a position-type signal based on the above equation (3 '). be introduced. 28 is an adding means, and 29 is a control target. Therefore, the FB control system is constituted by the above elements.

Next, the FB control system is combined with the FF control system.
The control system is configured as follows. That is, a disturbance amount detector 30 for detecting disturbances such as a production amount and a load is provided, and a coefficient means 31 is provided on an output side of the detector 30 to obtain a static FF control signal. Numeral 32 is a position type-speed type signal converting means for converting into a speed type signal which is a difference between the current value and the previous value of the static FF control signal, and 33 is a multiplying means.
The FF gain is corrected by multiplying the difference signal from the gain correction coefficient calculation means 34 by the gain correction coefficient obtained by the above equation (7), and the corrected gain is added to the addition means 26.

Next, the operation of the device configured as described above will be described. The process amount PVn from the control target 29 is detected by the process amount detector 21 and introduced into the deviation calculating means 23. The deviation calculating means 23 calculates the deviation en by using the target amount SVn and the process amount PVn from the target amount setting unit 22 to obtain the deviation en, and then calculates the deviation en by speed-type integral adjustment. Introduced to the calculating means 24, where the calculation is performed based on the above equation (3), and Ask for.

Thereafter, the speed-type operation change amount obtained by the calculation means 24 is calculated. Is supplied to the adding means 26, where it is added to the gain correction signal from the FF control system, and is added to the speed-position type signal converting means 27. The speed-position type signal conversion means 27 is based on the above equation (3 '). Position type manipulated variable M by integral operation And sends it to the gain correction coefficient calculating means 34 and the adding means 28. That is, here, the manipulated variable obtained by performing the calculation of the I component excluding the P component that changes rapidly Is used for the gain correction to the gain correction coefficient calculating means.

On the other hand, in the FF control system, the disturbance amount detector 30 detects a disturbance signal Dn due to a change in a production amount, a load, or the like, and calculates a coefficient means.
After calculating the static FF control signal FFn by the calculation of FFn = Dn · k using the coefficient k, the control signal FFn is converted to the position-speed signal conversion means 32 and the gain correction coefficient calculation. To the means 34 respectively. The position type-speed signal converting means 32 obtains a difference between the current value and the previous value of the static FF control signal, that is, a speed type signal of ΔFFn = FFn-FFn−1, and supplies it to the multiplying means 33. The gain correction coefficient calculating means 34 calculates the P component operation amount And the static FF control signal FFn, based on the above equation (7), By performing the following calculation, a gain correction coefficient kn-1 is obtained and introduced into the multiplication means 33. Here, the multiplication by kn-1 · △ FFn is performed to obtain the gain correction amount, and
Introduced to 26, gain correction by P calculation of FB control system.

Then, as described above, the speed-type operation change Is calculated by the speed-position-type signal conversion means 27 and supplied to the addition means 28. At this time, the position type proportional operation output of en · kp is input to the adding unit 28 using the deviation en and the proportional gain kp from the position type proportional adjustment operation unit 25, so that the output of the operation unit 25 and the signal The output of the conversion means 27 is added, and the result is applied to the control target 29 as an original operation amount for control.

Next, FIG. 2 is a diagram showing another embodiment of the apparatus of the present invention, in which the gain of both the FF control system and the FB control system is corrected. For example, the boiler steam pressure, steam flow rate, temperature ,
It is used for processes that involve changes in quality, such as concentration, while focusing on quantity. Specifically, the deviation calculating means 23
A multiplication means 40 is provided between the control means 24 and the speed-type integral adjustment calculation means 24, and the gain correction coefficient obtained by the gain correction calculation means 34 is provided.
kn-1 is sent not only to the multiplying means 33 but also to the multiplying means 40, and the gain of the FB control system is corrected by multiplying the deviation en by a gain correction coefficient kn-1.

Therefore, according to the configuration of the above-described embodiment, the P component adjustment operation and the I component adjustment operation are separated in the PI control, the I component operation is a speed type operation, and the P component operation is a position type operation. While respecting the essence of PI control
The advantage of the velocity type can be used while combining the FB control system and the FF control system, and the gain is corrected by the disturbance change using the I component except for the P component which has harmful fluctuation components, so that the disturbance change High-speed response, high-accuracy, and stable control can be executed by sufficiently coping with changes in process characteristics.

In the above embodiment, the differential adjustment operation is excluded from the FB control system, but may be added. Further, not only staticity compensation but also dynamic characteristic compensation may be added to the FF control system. In addition, the present invention can be implemented with various modifications without departing from the scope of the invention.

[Effects of the Invention] As described above in detail, the present invention has the following various effects.

First, with respect to claim (1), the P adjustment operation and I
Separate the adjustment operation from the I operation and perform the I
By making the adjustment calculation a position-type calculation, sufficient controllability and stability can be ensured even when the operation amount recovers after exceeding the upper and lower limit values, and there is no fluctuation component except for harmful fluctuation components. Since the gain correction coefficient is obtained using the signal, high-speed and high-precision gain correction can be performed.

Next, regarding claim (2), the FF control system and the FB
Since the gain of both the control system is corrected using the gain correction coefficient obtained from the average signal having no fluctuation component, it is possible to control the control object with improvement in quantity and quality.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of the device of the present invention, FIG. 2 is a block diagram showing a second embodiment of the device of the present invention, and FIG. 3 is an FF for obtaining a gain correction coefficient. FIG. 4 is a block diagram of a conventional device, and FIG. 5 is a diagram showing a relationship between an I component and a P + I component in a PI control output. 21 ... Plant quantity detector, 22 ... Target quantity setting unit, 23 ...
Deviation calculation means, 24 Speed-type adjustment calculation means, 25 Position-proportional adjustment calculation means, 26 Addition means, 27 Speed-
Position-type signal conversion means, 28 addition means, 29 control object, 30 disturbance amount detector, 31 coefficient means, 32 position-speed signal conversion means, 33 multiplication means, 34 ... Gain correction coefficient calculation means, 40... Multiplication means.

Claims (2)

(57) [Claims] 1. After a deviation is obtained by a deviation calculating means using a process amount and a target amount, a PI (P: proportional,
I: Integral) A control device in which a feedforward control system is added to a feedback control system that obtains an operation amount to control a controlled object by performing an operation, and performs a speed-type integral adjustment based on the deviation. Calculating means; a position-type proportional adjustment calculating means which is provided separately from the speed-type integral adjusting calculating means and performs a position-type proportional calculation based on the deviation; and a static feed-forward by multiplying a disturbance amount by a feed-forward coefficient. A first signal converting means for obtaining a control signal and converting the control signal into a speed-type signal which is a difference; and a position-type signal obtained by converting the output of the speed-type integral adjustment calculating means by a second signal converting means. Gain correction coefficient calculating means for calculating a gain correction coefficient using the signal and the static feedforward control signal; Compensating means for performing disturbance compensation by adding a static characteristic compensation signal obtained by multiplying the speed-form signal as the difference by the speed-correction coefficient to the output of the speed-form integral adjustment calculating means; and a signal compensated by the compensating means. Operation amount obtaining means for obtaining an operation amount signal for a control object by adding the position-type operation amount converted by the second signal conversion conversion means and the output of the position-type proportional adjustment operation means. A gain adaptive control device characterized by the above-mentioned. 2. A method according to claim 1, further comprising adding a multiplying means between said deviation calculating means and said speed-type integral adjustment calculating means, and applying a gain correction coefficient from said gain correction coefficient calculating means to said deviation. A gain adaptive control device characterized in that a feedback gain is corrected by multiplying by.