JPS5947609A - Process controller - Google Patents

Abstract

PURPOSE:To ensure good process control even through the static gain of a controlled system is nonliner, by obtaining a manipulated variable of a complement feedback controller. CONSTITUTION:A deviation arithmetic means 5 calculates the deviation value between the target value (r) and a manipulated variable (c), and this deviation value is applied to a coefficient element 3. The element 3 divides the deviation value by a static gain Kc which simulates the static characteristic of a controlled system 20 and delivers (r-c)/Kc. While a simulation deviation arithmetic means is provided within a process device 10 to simulate the dynamic characteristics of the controlled system 20. A differential amount arithmetic means 6 calculates the output of the element 3 and the output of the simulation deviation arithmetic means and delivers the differential amount between both outputs. This differential amount is integrated by an integration means 7 with the prescribed timing, and this output value controls the controlled syste 20 in the form of a manipulated variable (m). In such a way, the process control is satisfactorily performed even if the static gain of a controlled system is nonlinear.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] This invention relates to process control devices, and more particularly to control devices with complementary feedback loops.

[Prior art]

A ttjl'I diagram of a prior art control device with a complementary feedback loop is shown in FIG.

The process control device lO has a manipulated variable m for the controlled object 20.
Control this by outputting . The static characteristics and dynamic characteristics of the controlled object are expressed as static gain Kp and dynamic gain G, respectively.
p (s+).

- A static gain K that simulates the static gain Kp and the dynamic gain Gp(s) of the controlled object 20 also in the force control device 10.
c and dynamic gain Gc (s), and are configured to simulate the static characteristics and dynamic characteristics of the controlled object 20. The static gain Kc constitutes a coefficient element as a static characteristic part 3, and the dynamic gain Gc (s) constitutes a simulated deviation calculation means as a dynamic characteristic part V. The control amount C which is the output of the control device 1
It is configured to be fed back to the input terminal of 0 and calculate the deviation between it and the target value r. The deviation is input to the coefficient element 3, the output of the coefficient element 3 and the output of the simulated deviation calculation means are added, and f is set so that this becomes the manipulated variable m.
lt controller IO is configured.

If there is a deviation between the target value r and the controlled variable C, the manipulated variable m is outputted as m = (rc)/kC so that the deviation disappears. Here, Kc = Kp, Gc (
s) = Gp (s), even if the controlled variable C changes from the manipulated variable mVc and the deviation r-e decreases, the output of the simulated deviation calculation means μ increases by the amount of the decrease. Therefore, the manipulated variable m is kept unchanged.

In this way, if the static gain Kc and dynamic gain Gc (s) in the control device IO are set to satisfy the critical braking condition, a response similar to open circuit control can be obtained with the control amount being 0. It operates as an effective control device even for control objects that include dead time.

[Problems with conventional technology]

In this conventional process control device, when the static gain Kp of the process is nonlinear, it is
When the static gain Kc in the f control device is set to a non-6 type, the static gain Kc changes and corresponds to the L minute.
Also changes, and f and b deviate from the appropriate values.

Therefore, it takes a certain amount of time for the manipulated variable m to be corrected to the appropriate value. This has the disadvantage of poor controllability.

[Purpose of the invention]

An object of the present invention is to set the static gain Kp of the controlled object to a deviation r
-c vCJ: It is an object of the present invention to provide a process<i control device that has good ftj controllability even in a nonlinear control system where the number of changes increases or decreases.

[Summary of the invention]

In this invention, in order to achieve the above object, in order to calculate the manipulated variable m, the conventional method of adding the single lens guided by the deviation to the output of the machining process in the control device is changed, and the A method was adopted in which the difference amount between the amount of output and the value of υ derived from the simulated deviation by the simulated process in the fu control device i was multiplied by 3v. The deviation value from the control target value of the target is calculated from the static characteristics and dynamic characteristics of the target.
-! In a process control device that outputs a manipulated variable for the controlled object by using a static gain and a dynamic gain that simulate a a coefficient element that is combined and divides the deviation value by the static gain Kc to calculate a quantity derived from the deviation; an output value from this coefficient element; and an output value obtained by multiplying this output value by the dynamic gain Gc (s). and a means for calculating a difference between the amount guided by the deviation and the simulated deviation, and an integrating means for integrating the difference at a predetermined timing. The output of the integration means is configured to be the manipulated variable.

Embodiments of the present invention will be described in detail below with reference to the drawings.

[Embodiments of the invention]

FIG. 2 shows an embodiment of the present invention (1゛q diagram).

Note that the same parts as shown in FIG. 1 are given the same reference numerals and their explanations will be omitted.

The target value r and the control number C are manually input to the deviation value calculating means 5 to calculate the deviation value. This deviation <+(i, It") is manually entered into the coefficient element 3 coupled to this deviation value calculating means j.

What about coefficient element 3? This deviation value is divided by a static gain Kc that simulates the static characteristics of the Iil controlled object, and a quantity (r - c )/Kc derived from the deviation is output.

On the other hand, inside this process control device 10 [ha! fi controlled object 2
The zero difference calculation means 6, which is provided with a simulated deviation calculation means for simulating zero dynamic characteristics, calculates the output of the coefficient element 3 and the output of the simulated deviation calculation means, and outputs the difference amount.

- This amount of difference is accumulated in the product a means 7 at a predetermined timing, and the output value is used as the operation lm7 to control the controlled object unit O. Model JZi: In the embodiment shown in FIG. 2, the deviation calculation means are static gain (Kc, ) setting section r,
It consists of an integrating means, a QJ characteristic section, and a coefficient element/7. The output value from the coefficient element 3 is multiplied by the static gain Kc by the static gain setting section K and inputted to the dynamic characteristic section ≠ and the deviation calculation means /l via the integrating means. Deviation calculation means//- Calculates the deviation between the output of the calculation means and the output via the dynamic characteristic section μ and inputs it manually to the coefficient element /7. This coefficient ammonium water/7 has the same structure as coefficient element 3. The output of the coefficient element 3 and the output of the coefficient element 17 are configured to be manually input to the difference calculation means B.

Next, the operation of the control device shown in FIG. 2 will be explained by taking as an example a case where the target value r changes by Δr from a stable state where the target value r and the control amount c coincide with each other. Target value r is r+
When Δr is reached, the output of the deviation value calculation means j becomes Δr.

Therefore, the output of coefficient element 3 is Δr/Kc.

Since the output of the deviation calculation means l/ was stable at r=c, it is violet 0 at this time.

Therefore, the output of coefficient element /7 is also 0. The output of the differential degree calculation means t is therefore Δr/Kc.

This value is given to the integrating means 7, integrated with the previous integrated value, and output as the manipulated variable m. The control amount C gradually changes according to this operation amount m. Difference (11) The output of the operation means B is further multiplied by Kc by the static gain setting means g to become Δr, and the output of the Q integration means is input to the integration means. and
On the other hand, the output of the excited section is the response value of the simulated process.

Therefore, the output of the deviation value calculation means j is the deviation, whereas the output of the deviation calculation means // is the simulated deviation.

In this way, the control device i\10 repeats the above operations at predetermined timing. When control calculations are performed digitally at certain intervals, calculation errors can be reduced by setting the output of the dynamic characteristic section not to the value at the time of the calculation but to the value at the next calculation time.

If //Kc of coefficient elements 3 and 17 is linear, it becomes possible to omit the static gain setting means r, the integrating means, and the coefficient element 17 as shown in FIG. - In the control device, the manipulated variable m was obtained by adding the response value of the 4M-11'rli process set in the control device and the amount determined by the deviation at that time. Since it is determined by integration, even if the static gain Kp of the controlled object is nonlinear, it is possible to construct a control device that is not affected by nonlinearity.

- As an example, a process control device that continuously neutralizes inflow raw water (alkaline) will be explained.

In such a process control device, a method is used to control the chemical injection flow rate by calculating the chemical injection flow rate = raw water flow rate x injection rate, but the control device according to the present invention is used to determine the injection rate (d) Apply. Let the target water quality (pH 7) be the target value r, the process water quality (pH) be the controlled variable C, and the injection rate be the manipulated variable m.

The relationship between the water quality (pH) of raw water and the injection rate necessary to neutralize it is shown in Figure μ.

In a conventional complementary feedback control device, for example, even if the pH is determined to be A and the injection rate is determined to be A' at the start of control, when the control effect appears and the pH becomes B, the injection rate B' becomes the pH value.
Since A is smaller than the ratio of B, the necessary injection rate A' cannot be secured, and the target water quality pi
(The speed at which it approaches 7 is slowing down.

On the other hand, in the control according to the present invention (ii), if the pH is determined to be A and the injection rate is determined to be A' at the start of the 1-period control, from then on, the system is set up in the device to simulate the process and to prevent counterfeiting. As long as there is no approximation error between the injection rate and the process, the injection rate is maintained at A', so the speed at which the target water quality (pH 7) is reached is faster than that of the conventional complementary feedback control device.

FIG. 3 shows another embodiment of the present invention and is a diagram 4'II. The difference from the embodiment shown in Fig. 2 is that Fig. 2 calculates the correction amount of the manipulated variable m using the coefficient element of / / Kc for the deviation and predicted deviation. The two combinations in Fig. 3 include the target value r and '+(i control r'#, c, and 1 for each of the target value and output value of the dynamic characteristic section Hiroshi.
The point is that the correction amount of the manipulated variable m is obtained using the different types of polygonal line elements. Polyline element/Yu, 13. /4'.

lS gives the relationship between the water quality (pH) of raw water and the injection rate necessary for its neutralization, as shown in Figure μ, for example. On the contrary, the reverse polygonal line element /A gives the water quality (pH) of the raw water to the injection rate.

The coefficient element / /Kc shown in Fig. 2 is a conversion element similar to this polygonal line element, but since the coefficient element / /Kc is for deviation, it can be changed even if it is i difference. If r changes, its value must be changed.

In contrast, in the embodiment shown in FIG. 3, the same polygonal line element can be applied even if the target value r changes. Note that the prayer line element is an approximation of the relationship between the vertical axis and the horizontal axis shown in Figure 1 using a polygonal line, but it is also possible to replace it with a mathematical formula without using such an approximation formula.

〔Effect of the invention〕

As described above in detail based on the embodiments, the present invention uses the operation amount to make the deviation between the target value and the controlled amount zero, and the one-time effort to make the simulated deviation in the simulated process built into the control device O. Since the operation finger of the controlled object is combined with the operation finger of the controlled object and the difference value is integrated and outputted, the control response is well linked even when the static gain Kp of the controlled object is non-linear. Process control with controllability 1
There is an added advantage of being able to arrange a noni storage.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a conventional complementary feedback control device, FIG. 2 is a block diagram showing a conventional complementary feedback control device, FIG. , an example in which the static gain of the controlled object is linear, Fig. It is a block diagram showing another example of the invention. 3... Coefficient element, Hiroshi... Dynamic characteristic section, j... Deviation value calculation means, t... Difference amount calculation means, 7. ...Integration means.g...Static gain setting means, /7...Coefficient element, /
/... A 1 ml difference calculation means, 10... Process control device, 20... Controlled object, r... Target value, C...
Control amount, m...operation 4B.

Claims (1)

[Claims] The deviation value between the control amount fed back from the controlled object and the control target value of the controlled object is determined by a static gain Kc and a dynamic gain Gc (s
), the process control device outputs a manipulated variable for the controlled object, and includes a deviation value calculation means for calculating the deviation value, and a deviation value calculation means coupled to the deviation value calculation means to calculate the deviation value to the static gain Kc. a coefficient element for calculating a quantity derived by the deviation by dividing by , and a simulated deviation for calculating a simulated deviation from an output value from this coefficient element and an output value obtained by multiplying this output value by the dynamic gain Gc (s). comprising a calculating means, a means for calculating a difference amount between the amount derived by the deviation and the simulated deviation, and an integrating means for integrating the difference amount at a predetermined timing; A process control device characterized by: