JPH0195301A - Self-tuning controller - Google Patents

Abstract

PURPOSE:To facilitate the operation of a self-tuning controller by superposing a step signal onto an operation signal which turns on a pointing means and therefore attaining a tuning action. CONSTITUTION:When an operator operates a pointing means 4, a step type signal is superposed on an operation signal and outputted to a controlled system. In this case, a parameter tuning means 3 observes the produced process value or the waveform of a deviation signal and tunes a PI arithmetic parameter. Thus a user can easily perform a tuning action as necessary with operation of the means 4.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a self-tuning controller that automatically adjusts at least proportional (P) and integral (I) calculation parameters to an optimal one-shift. This invention relates to a Hilf tuning controller that determines P and I calculation parameters by recognizing the waveform of disturbances such as randomly occurring disturbances, without using special signals to identify processes.

(Prior art) BuOt? used for feedback control? A self-tuning controller that automatically tunes the PI parameter has been put into practical use as a system controller. Among these, a pattern recognition means is provided to observe the behavior of the control m, and this recognizes disturbances in the control system such as randomly occurring disturbances without causing any disturbance to the process, and this recognition result is used to optimize the In comparison with a response model, a method in which the In 3m PII stop parameter is determined so that the pattern recognition result approaches the response model is attracting attention.

For this kind of self-tuning controller, the book is
ADA,,PTIVE C0NTR0L SYST
EMS, PERGAMON PRESS.

1963 R'fTP1-P18 and US PATE
NT No. 4.602.326 is known.

(Problem to be solved by the invention) By the way, such a self-tuning controller is
Since it is configured to tune the PI calculation parameter using disturbances in the wJ system, for example, in a process of stable nature (a process in which disturbances occur only rarely), the P calculation parameter may be slightly adjusted. Even if the value is sloppy, the control is stable, so there is no opportunity to tune the PI calculation parameters.

For this reason, there may be cases where it is desired to tune the optimal PI calculation parameters in advance in case disturbances, fluctuations, etc. occur in the future.

An object of the present invention is to realize a self-tuning controller that meets these demands.

(Means for Solving the Problems) FIG. 1 is a basic functional block diagram of the apparatus of the present invention. In the figure, 1 is a controlled object (process) whose dynamic characteristics change due to changes in production I, changes in control target values, disturbances, etc. 2 is the process mP from the controlled object
The operation signal MV obtained by performing at least proportional (P) and integral (I) calculations on the deviation signal DV between the control port gmsv and
The PI control means 3 outputs P[1! II
This is a parameter tuning means for tuning the P and I calculation parameters of the m means. Reference numeral 4 designates an instruction means for externally instructing a tuning operation, and reference numeral 5 designates a step signal generation means for receiving a signal from the instruction means 4 and superimposing a step-like signal on the operation signal MV.

(Function) For example, when an operator operates the instruction means, a step-like signal is superimposed on the operation signal and output to the controlled object. The parameter tuning means converts the waveform of the process signal or deviation signal generated at that time into I! and tune the PI calculation parameters.

(Example) Embodiments of the present invention will be described in detail below using the drawings.

FIG. 2 is a block diagram of an embodiment of the present invention.

In the figure, 6 is a self-tuning controller, and 1
1J111 Process IPV from object 1.

The control target msv is input, and the operation signal MV is output to the controlled object 1. In this controller, numeral 61 is a multi-preflexor that sequentially selects and inputs the process mpv, various analog signals ei, and control port mm5v, and 62 is a comparator that inputs the signals selected by the multipreneur 61 into one input.
A microprocessor 63 inputs signals from the comparator 62 and the like. 64 is a D/A converter that converts the digital signal from the processor into an analog signal; 65
is a sample and hold circuit that holds the output of the D/A converter at a predetermined timing, and its output becomes the operation signal @MV to the control device 1. 66 is an I10 boat, and 67 is a RAM that stores various data.

-68 is a ROM 169 that stores programs for main operations performed by the microprocessor 63; a ROM 169 that stores programs created by the user according to the process;
are display keyboards which are coupled to microprocessor 63 via data bus 88. An instruction means 4 for instructing a tuning operation from the outside is provided in a part of this display keyboard.

Here, the system ROM6B contains MicroPro Tuna 6
3 stores programs for performing the functions of the PI control means 2, parameter tuning means 3, and step signal generation means 5 shown in FIG. 1, and data of the 1&appropriate response model of the control pair @1. In addition, system ROM6
8. The ROM 69 may be shared by one ROM.

FIG. 3 is a flowchart showing an example of the operation of the apparatus configured as described above.

MicroProcera+J'63 first has a system ROM
The process mpv, control target value S, and other analog signals ei applied to the multiplexer 61 are sequentially selected and taken out according to the system block diagram stored in the system block 68, and these signals are sent to the comparator 62 and the microprocessor 63. , are converted into digital signals by an A/D conversion loop formed by a D/A converter 64 (
Step 1).

Next is PuOse, 1iPV! IJ ill It is determined whether the deviation D■ from the target msv is within a predetermined set value range DB (step 2). When 't'ov>os, the process moves to the next step 3. In step 3, waveform observation of the process ff1PV or deviation DV is performed according to the system program functioning as the parameter tuning means 3 in FIG. 1, and a pattern analysis of the observed waveform is performed. When the pattern analysis is completed, the evaluation index of the l1i11 waveform is calculated, the PI calculation parameters are calculated based on the evaluation index, and the PI calculation parameters are set (tuned) in the PI control means (step 4.5.6 )
.

Next, the P1tlJtlA means 2 performs PI calculation using the P11 parameters set and changed by the parameter tuning means 3 (step 7), and outputs the calculation result to the controlled object 1 as an operation signal MV (step 8).

In step 2, when DV<DB (when the deviation DV is within a predetermined set value range), it is determined that the control is good and the process moves to step 21. In this step 21, it is determined whether the instruction means 4 is on or not (or whether the instruction means 4 is operated).

If the determination is No here, the process moves to step 8. Y
If it is determined that ES is the case, the step signal generating means 5 superimposes a step signal on the operation signal MV (step 2).
2).

Next, the parameter tuning means 3 processes the process ff1
Observe the waveform of PV or deviation DB, analyze the pattern (Step 23), and perform PI based on this analysis result.
Compute the calculation parameters and set the PI performance parameters of P
I control means 2 (step 24).

Through the operations of steps 22 to 24 described above, even if the control is in a good state, the optimum PI fraud parameters can be tuned in advance by operating the instruction means 4 in preparation for future disturbances, fluctuations, etc. .

FIG. 4 is a time chart showing an example of the operation. T1
When the instruction means 4 is turned on at the timing of (a)
The step signal MO is superimposed on the operation signal MV as shown in FIG. Here, the magnitude and direction of the step signal MO (polarity)
shall be determined in advance based on the characteristics of the controlled object 1.

Further, this step signal is generated, for example, at timing T1, when the PI control means 2 determines, when MC>O, integral term←integral term calculated value++? When 1c 1MO<O, it is possible to output by performing a calculation such as integral term←integral term calculated value−IMcl.

When the operation signal MV is superimposed with Sfufu No. 13 Fukuyaku-Jin\, a fluctuating waveform as shown in (b) is observed in the process mpv. The parameter tuning means 3 observes this fluctuating waveform, performs a waveform analysis, completes this waveform analysis at timing T2, calculates a PI calculation parameter, and sets it in the PI control hand fR2.

If further tuning is required as a result of waveform T2 measurement at timing 2, waveform observation is continued and tuning is performed until the deviation is small and stable. When the above operations are completed, the state returns to the state before the tuning request.

Note that in the above example, the step signal is output only once when a tuning request is made, but the fifth step signal is output only once.
As shown in the figure, a step signal may be generated several times with one tuning request.

FIG. 6 is an explanatory diagram showing an example of a waveform observation method in the parameter tuning means 3.

This shows the response waveform of the deviation signal after applying a step signal, and although it fluctuates greatly at the beginning, the PI
Due to the operation of the IJm means 2, the change gradually becomes smaller and eventually converges to zero.

The parameter tuning means 3 observes the behavior of this deviation signal Dv, and the first peak ([
) Vl) was observed (time of occurrence of the first peak) t
The area A1 related to the deviation signal from 1 to the time point t2 at which the second peak (DV2> is observed (the time point at which the second peak occurs)) and t2, which continues from the time point t2 at which the second peak occurs.
+(t2-tl) (a point after t2-tl has elapsed from t2), the area A2 related to the deviation signal Dv is calculated. And these areas A1. Using A2, A
1. Calculate the A2 ratio (AR).

Area A1. The formula for calculating A2 and area ratio AR is (I),
These are shown in equations (2) and (3), respectively.

△1=J″″ (DVl-DV) d t
1 tongue I Here, the calculation formula of equation (I) is to find the area A1 shaded in FIG. 6, and the calculation formula of equation (2) is to calculate the area Δ2 shaded in FIG. This is what we seek.

In addition, the maximum overshoot layer (overshoot) OVS is (4
) is calculated using the arithmetic expression of Eq.

0VS=-DV2/DV1 4Furthermore, the vibration period Tp is determined by Equation 8Iiii of Equation (5).

Tp=2・<t2−tl) 5 This information is used as an evaluation index for the observed waveform, such as the area ratio AR and overshoot period Tp, which are obtained using Tune the PI calculation parameters.

Note that if the area ratio AR is used as one of the evaluation indicators, it has the characteristic that it is not influenced by noise included in the observed waveform.

(Effects of the Invention) As described above in detail, according to the present invention, when the user feels that tuning is necessary, for example, the user can easily perform the tuning operation by operating the instruction means, and Self-tuning controllers can be used with confidence even in processes where problems occur only rarely.

[Brief explanation of the drawing]

Fig. 1 is a basic functional block diagram of the device of the present invention, Fig. 2 is a block diagram of the configuration of an embodiment of the invention, Fig. 3 is a flowchart showing an example of the operation of the device of the present invention, and Fig. 4 is the operation. 5 is a time chart showing an example of another operation, and FIG. 6 is an explanatory diagram showing an example of waveform observation/method in the parameter tuning means 3. 1...Controlled object 2...PI control means 3...Parameter tuning means 4...Instruction means 5...Step signal generation means Fig. 1 Fig. 2 Fig. 3 Fig. 2, ', 6 Q

Claims (1)

[Scope of Claims] PI control means for performing at least proportional (P) and integral (I) calculations on the deviation signal between the process amount from the controlled object and the control target value and outputting the obtained operation signal to the controlled object; , a controller comprising parameter tuning means for observing the waveform of the process quantity or deviation signal and tuning the P and I calculation parameters of the PI control means so that the observed result becomes a preset response target. . A self-tuning controller comprising: instruction means for externally instructing a tuning operation; and step signal generation means for receiving a signal from the instruction means and superimposing a step-like signal on the operation signal.