JPH04107601A - Process identifying device - Google Patents

Abstract

PURPOSE:To save the time and the electric power needed for identification of a process by putting a process identifying process into a process rising process. CONSTITUTION:A control output changing means 3 instructs an output computing part 14 to output the maximum control output MV to a process 2 and then reduces the output MV by a prescribed amount when the differential operation value D received from a differential operation circuit 13 is stabilized. A time counter means 5 counts continuously the time TL until the change of the output MV affects the value D, that is, a differential operation output change detection means 4 detects the change of the output MV after occurrence of this change. An arithmetic parameter setting part 6 identifies the time TL as the equivalent dead time of the process 2 and calculates a control arithmetic parameter based on the identifying result. Then the part 6 sets this arithmetic parameter at a PID control arithmetic part 1. Thus the process 2 can be identified by means of the rise of the process 2 and therefore the time and the electric power needed for identification of the process 2 can be saved.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a process identification device for identifying dynamic characteristics of a process, and more specifically, to a process identification device for identifying the dynamic characteristics of a process. The present invention relates to a process identification device used for automatically setting calculation parameters (proportional band PB, integral time constant TI, differential time constant TD, etc.).

<Prior art> In order to control the process to be controlled under I&appropriate conditions,
A process identification device is required to know the dynamic characteristics of the process.

There are two types of process identification devices: one where a controller that controls the process is connected to the process and identification is performed in a closed loop while controlling the process, and the other where the controller is separated from the process and identification is performed in an open loop. .

FIG. 3 is a block diagram showing a control system for boat fishing, and describes a conventional example in which identification is performed in a closed loop while controlling 10 cells.

The PID control calculation unit (main calculation unit) 1 inputs the target set value SP and the process amount Pv from the process 2, and calculates the deviation D.
Proportional (P), integral (I), and differential (D) calculations are performed on V, and the calculation results are output to process 2 as control output (manipulated amount) MV, so that the process amount Pv becomes the target @ set value SP. Control to match.

The process identification unit 3 regulates the operation of the PID control calculation unit 1, performs on/off control of the process 2, monitors the process amount Pv at that time, measures the dynamic characteristics of the process, and uses the results as The optimum PID calculation parameters (calculation constants) are calculated based on the PID control calculation unit 1.
Automatically set to .

FIG. 4 is a waveform diagram showing an example of a change in the process amount PV observed during the identification operation by the process identification section 3.

(a) to (c) all show different processes. The process identification unit 3 determines the vibration amplitude and vibration period of the process amount PV, and determines the optimal PID based on these values.
Calculation parameters (calculation constants) are calculated and automatically set in the PID control calculation section 1.

<Problems to be Solved by the Invention> However, with such conventional process identification methods, when starting up a batch furnace, for example, a trial run is first performed on an empty furnace for process identification, and then a test run is performed on an empty furnace for process identification. It is necessary to take steps to introduce the actual material to be heat treated and conduct actual operation.

This has the problem that not only is a trial run step required for process identification, but also that power is wasted in that step.

The present invention has been made in view of these points, and
The purpose is to realize an apparatus that can save time and power related to process identification by incorporating a process identification step into the process start-up process.

Means for Solving Problems> The present invention that achieves the above objects comprises a control output changing means for changing the control output output to the process, and a differential operation for a signal related to the process amount from the process. The differential calculation circuit is configured to include a differential calculation circuit, and a time measurement unit that measures the time (TL) from when the control output is changed by the control output changing unit until the effect thereof appears in the differential calculation amount of the differential calculation circuit. Ru.

<Operation> The control output changing means waits until the calculation output from the differential calculation circuit becomes stable, and then performs a changing operation to temporarily reduce the control output.

The time measuring means measures the time (TL) from when the control output is changed by the control output changing means until the effect thereof appears in the differential calculation amount from the differential calculation circuit. Then, the measured time (TL) is identified as the equivalent dead time of the process, and control calculation parameters are calculated based on this identification result.

<Example> Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 1 is a block diagram showing an embodiment of the present invention.

In the figure, 1 is a PID control calculation unit that inputs the target set value SP and the process amount Pv from the process 2, performs PID calculation, and outputs the control output (manipulated amount) MV to the process 2. A proportional calculation circuit 11 which inputs the deviation signal DV between SP and the process quantity Pv, an integral calculation circuit 12, and a differential calculation quantity F#11 which inputs the process quantity P■.
3 and an output calculation unit 14 that adds the calculation results from each of these calculation circuits and outputs the result to the process 2 as a control output MV.
It is made up of.

Reference numeral 3 denotes a control output changing means for changing the control output MV given to the process 2, which performs an operation such as reducing the control output given to the process by a predetermined amount, for example.

4 is a differential calculation output change detection means for detecting a change in the differential calculation output from the differential calculation circuit 13; 5 is a time measurement means;
Time (TL) from when the control output MV is changed by the control output changing means 3 until the effect appears on the differential calculation amount from the differential calculation circuit (until the change is detected by the differential calculation output change detection means 4) Measure.

6 is a calculation for identifying the time (TL) measured by the time measuring means 5 as the equivalent dead time of the process, calculating control calculation parameters based on this identification result, and setting them in the PID control calculation unit 1. This is a parameter setting section.

Here, the calculation parameter setting section 6 is configured to calculate the PID calculation parameter using the time (TL) from when the control output is changed until the influence appears in the differential calculation amount of the differential calculation circuit 13. has been done.

The operation of the apparatus configured in this way will be explained next.

FIG. 2 is a waveform diagram showing the operation of process identification.

When identifying a process, first receiving instructions from the operator, the control output changing means 3 acts on the output calculation section 14, and from there the maximum output (100%) as shown in (a) is output.
) is output to process 2. In response to such a control output MV, a process amount (step response signal) PV toward the target setting value SP is obtained from the process 2 as shown in (b).

The differential calculation circuit 13 receives such a change in the process amount Pv and calculates the differential calculation amount (D). (a) shows the change in the differential calculation amount (D).

The control output changing means 3 performs a changing operation to reduce the control output difference MV by a predetermined amount (for example, 50%) at a time t1 when the differential calculation amount (D) from the differential calculation circuit 13 becomes stable.

The process 2 subjected to this operation is affected to some extent, and the effect eventually appears as a change in the process amount Pv, which in turn appears as a change in the differential calculation output of the differential calculation circuit 13.

The differential calculation output change detection hand Pi4 detects a change in the differential calculation amount (D>) from the differential calculation circuit 13, and the differential calculation amount (D) changes rapidly due to the influence of a change in the control output. The zero time measuring means 5 that captures the time t2 has a control output M
The time TL from the time t1 when a change operation is made to V until the change time t2 is detected, and the measured time TL is transmitted to the calculation parameter setting section 6.

The calculation parameter setting unit 6 calculates this measurement time (TL) and
Amount of change ΔP in process amount PV during this measurement time TL
Using V, the calculation shown in equation (1) is performed to find the proportional band PB.

In addition, let the integral time constant TI and the differential time constant TD be, for example, (
2) and (3) are used, and these calculation constants are set in the PIDilJal calculation unit 1.

PB=APV (%/sec)*TL(sec)-(1)TI=
2 to TL*...(2) TD=T
3 to L* (3) After that, for process 2, the control output MV is calculated based on the above-mentioned process identification result, and the control output MV is calculated based on the set new 1NK parameter, and the process amount Pv is From approximately the middle point, the temperature will settle smoothly toward the target set value SP.

Such a process identification method can be performed using the start-up of a process, and can save time and power for process identification.

In addition, in such a process identification method, the change that appears in the process amount PV due to the operation of reducing the control output MV is, for example, a value of 0.1%, 0.2%, 1%, or 2% precision. , there is no influence on the actual process, and there is almost no time loss until the process amount PV reaches the target setting value SP. In the embodiment, the control output changing means 3. Although the differential calculation output change detection means 41 and the time measurement means 5 are shown as separate programs, each of these functions is executed by, for example, a microprocessor program, so it is arranged so that they can be executed by a single program. Good too.

Further, the PID control calculation unit 1 may be a PI control calculation unit.

<Effects of the Invention> As described in detail above, according to the present invention, a process can be identified during process start-up, thereby saving time and power.

Further, by calculating calculation parameters based on the identification results obtained by this method, a controller with excellent controllability can be provided.

[Brief explanation of the drawing]

FIG. 1 is a configuration block diagram showing one embodiment of the present invention, and FIG.
Figure 3 is a waveform diagram showing the operation of the process identification means, Figure 3 is a block diagram representing a general control system, and Figure 4 shows an example of changes in process quantities observed during identification operation by the process identification unit. FIG. 5... Time measuring means, 6... Calculation parameter setting section, 11... Proportional calculation circuit, 12... Integral calculation circuit, 1
3... Differential calculation circuit, 14... Output calculation section, 1...
- PID control calculation unit, 2... process, 3... control output changing means, 4.
・Differential calculation output change detection means,

Claims (2)

[Claims] (1) A control output changing means for changing the control output output to the process; a differential operation circuit for performing a differential operation on a signal related to a process amount from the process; and after changing the control output by the control output changing means. , and a time measuring means for measuring the time (TL) until the influence appears on the differential calculation amount of the differential calculation circuit, and after the control output is changed, the influence appears on the differential calculation amount of the differential calculation circuit. A process identification device characterized in that a time until appearance (TL) is identified as an equivalent waste time of a process. (2) A claim characterized in that the proportional band PB is calculated according to the following formula using the time (TL) from when the control output is changed until the effect appears in the differential calculation amount of the differential calculation circuit. Item 1. Process identification device according to item 1. PB=ΔPV(%/sec)*TL(sec)...(1)