JPH0424802A - Process controller - Google Patents

Abstract

PURPOSE:To ensure the proper control of the constant value without producing the large overshoot despite a control deviation caused by the disturbance, etc., by correcting an integral manipulated variable with the correction value decided by a fuzzy inference type integral manipulated variable correction means. CONSTITUTION:A fuzzy inference part 9 consists of a fuzzy computer or a microcomputer and performs a fuzzy inference based on the given control deviation (e) and manipulated variable U. Then the part 9 outputs the integral correction value to an adding point 19 as the result of the fuzzy inference. At the point 19 the integral value of the deviation (e) given from an integrator 17 is subtracted by the integral manipulated variable correction value given from the part 9. Thus the integral value given to an integral manipulated variable arithmetic means 5 is corrected in response to the deviation (e) and the variable U. Thus it is possible to obtain the excellent control characteristic having a short adjustment time, a high response speed, and the small overshoot. Furthermore the adjustment of the gain is facilitated.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention relates to a control device that performs process control such as temperature adjustment control, and in particular outputs a manipulated variable including an integral manipulated variable, such as PI control and PID control. The present invention relates to a process control device.

(Prior Art) As a process control device that performs process control such as temperature adjustment control, a PI control device or a PID control device uses a control device based on a control deviation between an arbitrarily determined control target value and a controlled amount from a controlled object. 2. Description of the Related Art A type of process control apparatus that outputs a manipulated variable including an integral manipulated variable in addition to a proportional manipulated variable to a controlled object is well known.

This type of process control device has linear control characteristics,
It is used for constant value control to keep temperature etc. at a predetermined control target value.

In the process control equipment as mentioned above, control safety,
Depending on the capability of the operating device, there is an upper limit value and a lower limit value for the manipulated variable given to the controlled object, and when the manipulated variable as a result of the calculation is about to change beyond the manipulated variable upper limit value or the manipulated variable lower limit value, That is, when the manipulated variable is saturated, further changes in the manipulated variable are suppressed.

The method of suppressing the change in the manipulated variable when the manipulated variable is saturated is that when the manipulated variable is in the saturated region, the direction of change in the integral manipulated variable and the total of the manipulated variables other than the integral manipulated variable are set to the upper limit of the manipulated variable,
Or find out whether the lower limit of the manipulated variable has been reached, etc.
It has been considered to stop the integration depending on these situations to prevent the occurrence of an overload phenomenon in the integration operation.

This is a patent application filed in 1986-3 by the same applicant as the applicant.
This method has already been proposed in No. 32112 (Japanese Unexamined Patent Publication No. 1-173201).

(Problem to be Solved by the Invention) The above-mentioned modification of the integral manipulated variable is limited to integration under saturation, and only the saturation of the manipulated variable is observed and execution or non-execution of integration is simply switched. Although the overshoot of the target value response is improved, if a control deviation occurs due to a disturbance or the like, the correction function does not work properly and the occurrence of overshoot is unavoidable.

SUMMARY OF THE INVENTION In view of the above-mentioned problems, it is an object of the present invention to provide an improved process control device that can perform appropriate fixed-value control without causing large overshoot even when a control deviation occurs due to disturbance or the like. .

(Means for Solving the Problem) According to the present invention, the above object is to provide a process control device that outputs a manipulated variable including an integral manipulated variable based on a control deviation between a control target value and a controlled variable. further comprising a fuzzy inference type integral manipulated variable correction means for determining a modification amount of the integral manipulated variable by fuzzy inference using the control deviation and the manipulated variable as input requirements; This is achieved by a process control device characterized in that it is configured to be corrected by the correction amount determined by the means.

(Operation) According to the above-described configuration, the integral manipulated variable is also corrected in the manipulated variable saturation control section using the correction amount determined by fuzzy inference based on the control deviation and the manipulated variable.

(Example) The present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 shows the basic configuration of a process control device according to the present invention. The process control device according to the present invention includes a PID
It has a control section 1 and a fuzzy inference section 9 for modifying the integral operation amount.

The PID control section 1 may have a well-known structure, and includes a proportional operation amount calculation means 3, an integral operation amount calculation means 5, and a differential operation amount calculation means 7.

The proportional manipulated variable calculation means 3 is given a control deviation e between the controlled variable Y from the controlled object (plant) 11 and the control target value R from a summation point 13, and determines the proportional manipulated variable based on this control deviation e. This is then output to a summing point 15.

The integral operation amount calculation means 5 is given an integral value of the control deviation e that is integrated by the integrator 17 and corrected at the summing point 19 in a manner described later, and calculates the integral operation amount based on this integral value. This is then output to a summing point 15.

The differential operation amount calculation means 7 receives the differential value of the control deviation e from the differentiator 21 and calculates the differential operation amount based on this,
This is output to a summing point 15.

The addition point 15 adds together the proportional operation amount from the proportional operation amount calculation means 3, the differential operation amount from the differential operation amount calculation means 5, and the integral operation amount from the integral operation amount calculation means 7,
A manipulated variable U consisting of the total value is output to the manipulated variable saturation control section 23.

The manipulated variable saturation control unit 23 suppresses the manipulated variable U from increasing or decreasing beyond a predetermined manipulated variable upper limit value or manipulated variable lower limit value, and sets an upper limit on the manipulated variable output to the controlled object 11. It is designed to fit between the value and the lower limit value.

The fuzzy reasoning section 9 is constituted by a fuzzy computer or a microcomputer, and is given the control deviation e and the manipulated variable U, and uses the control deviation e and the manipulated variable U as the human power requirement, as shown in FIG. Fuzzy inference is performed according to fuzzy rules, and the integral correction amount is output to the summation point 19 as the fuzzy inference result.

In the fuzzy rule shown in FIG. 2, NBSNMSSNS and ZRSPS are used as language information.

PM, PB are used, NB is negatively large, NM
is negative and medium, NS is negative and small, ZR is approximately zero, P
S is a fuzzy label indicating that it is positively small, PM is positive and medium, and PB is positively large.

The integral operation amount correction amount is determined according to the fuzzy rule as described above, and the integral operation is performed by subtracting the integral value of the control deviation e from the integrator 17 at the summing point 19 by the integral operation amount correction amount from the fuzzy inference section 9. The integral value given to the quantity calculating means 5 is corrected according to the control deviation e and the manipulated variable U.

As a result, when the control deviation e is small and the manipulated variable U is large, the output of the fuzzy inference section 9, that is, the integral manipulated variable correction amount increases according to PSSPMSPB and its degree, and accordingly, the integral manipulated variable calculation means 5 The integral value given to becomes smaller, and the integral manipulated variable calculated by the integral manipulated variable calculation means 5 correspondingly becomes smaller. This suppresses the occurrence of overshoot.

On the other hand, when the control deviation e is large, such as at the start of operation, the fuzzy output, which is the integral manipulated variable correction amount, is NS,
NMSNB decreases according to its degree, thereby increasing the integral value given to the integral manipulated variable calculating means 5, and as a result, equivalently, the gain of the integral manipulated variable calculating means 5 increases and the response becomes faster. .

As a result, even if the gains of the integral manipulated variable calculating means 5 and the differential manipulated variable calculating means 7 are not set strictly appropriately, excellent control characteristics with short settling time and small overshoot can be obtained. .

Figures 3 and 4 show the control amount and operation in the case where the integral value is corrected by the correction amount determined by fuzzy inference as described above, and in the case where only unintegration is performed when the operation amount is saturated. It shows the change in amount over time. From this graph, it can be seen that when the integral value is corrected by the amount of correction determined by fuzzy inference, control characteristics with shorter settling time and less overshoot can be obtained compared to integration under unsaturated manipulated variable. It will be.

In the above embodiment, the control deviation and the manipulated variable are used as the human power requirements for the fuzzy inference for modifying the integral manipulated variable, but in addition to this, the differential value of the control deviation is another input. A fuzzy rule may be set as a three-condition-conclusion. In this case, if a fuzzy rule is set so that when the differential value of the control deviation is large, the integral manipulated variable correction amount becomes small, and conversely, when the control deviation is small, the integral manipulated variable correction amount becomes large. good.

(Effects of the Invention) As understood from the above explanation, according to the process control device of the present invention, the integral manipulated variable is adjusted to the manipulated variable saturation control section with the correction amount determined by fuzzy reasoning based on the control deviation and the manipulated variable. This also provides excellent control characteristics with short settling time, fast response speed, and small overshoot, and facilitates gain adjustment.

[Brief explanation of drawings]

FIG. 1 is a schematic configuration diagram showing one embodiment of a process control device according to the present invention, FIG. 2 is a rule table diagram showing fuzzy rules for fuzzy inference used in the process control device according to the present invention, and FIGS. FIG. 4 is a graph showing the temporal change characteristics of the controlled variable and the manipulated variable, respectively. 1...PID control device 9...Fuzzy inference unit 23...Manipulated amount saturation control unit Patent applicant: Oromun Co., Ltd.

Claims (1)

[Claims] 1. In a process control device that outputs a manipulated variable including an integral manipulated variable based on a control deviation between a control target value and a controlled variable, the integral manipulated variable is determined by fuzzy inference using the control deviation and the manipulated variable as input requirements. further comprising a fuzzy inference type integral operation amount correction means for determining a correction amount of the fuzzy inference type integral operation amount correction means, and configured such that the integral operation amount is corrected by the correction amount determined by the fuzzy inference type integral operation amount correction means. Process control equipment.