JPH06337703A - Control device - Google Patents

Abstract

(57) [Abstract] [Purpose] The present invention can easily determine a control cycle from the sampling cycle of an analyzer, etc., and can secure an appropriate operation by a proportional term, which is high even for a control system with a long dead time. It is to exert controllability. [Configuration] In a control device that executes a calculation using a predetermined control constant based on a deviation that is a difference between a target value and a control amount, and calculates an operation amount to be applied to a process, a control amount such as an analyzer The control cycle calculation means 32 for determining the control cycle from the changing sampling cycle of the measuring device, the control constant predetermined based on the deviation between the target value and the control amount, and the control cycle calculated by the control cycle calculation means Integral term calculation means 33 is used to calculate the integral term calculation output to obtain the integral term calculation output, proportional term calculation means 34 is used to perform the proportional term calculation using the difference in the control amount, and both calculation outputs of these calculation means. This is a control device provided with an adding means 35 for adding to obtain an operation amount.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device used in various instrumentation control systems for monitoring and controlling plant conditions such as chemical plants including petroleum industry plants, steel plants, etc. The present invention relates to a control device provided with a technique for improving the controllability of a control system.

[0002]

2. Description of the Related Art Conventionally, various component controls including concentration control have been used in chemical plants and the like. However, for example, component control of a distillation column is a control system with a long dead time and a long response. There is.

In such a control system, a so-called sample PI control system is widely used in which PI (P: proportional, I: integral) control is executed while intermittently sampling the plant state. This is because if the normal PID (D: differential) control is applied to a control system with a long dead time, the manipulated variable is updated one after another before the effect of the manipulated variable on the process is confirmed, resulting in a sufficient control effect. Is not expected.

Therefore, in the sample PI control, since the basics of the feedback control are faithfully executed, the next manipulated variable is calculated after the effect of the manipulated variable sufficiently appears, and the calculated manipulated variable is applied to the process. The method to do is. That is, as shown in FIG. 5, the operation of the sample PI control is performed P for the control execution time t ₁ at every predetermined cycle t _0.
In this method, the I control is executed, and thereafter, the manipulated variable MV as the execution result is held constant. The control formula at this time is as follows.

[0005] _{△ MV n = K P (e} n -e n-1) + (τ / T I) e n ...... (1) MV n = MV n-1 + △ MV n However, control stop of the equation The inside is set to e _n = 0. Also, △ M
V _n : amount of change in manipulated variable, K _P : proportional gain, e _n : deviation at current time n, e _n-1 : deviation at previous time _n-1 , T _I : integration time, τ: control cycle, MV _n : Current manipulated variable, MV
_n-1 : Previous manipulated variable.

FIG. 5B shows the control characteristics when K _P ≠ 0. When K _P ≠ 0, it appears as a protruding state based on the fact that the control deviation e _n = 0 at the end of t ₁ , that is, after the elapse of the dead time.

The sample PI control as described above is intended for the purpose of controlling a control system having a large dead time or time constant, but the control execution time and the control cycle are always constant. However, in actual plant control (for example, component control of a distillation column), the transfer time changes depending on the size of the raw material flow, and the dead time changes accordingly, so it is possible to flexibly respond to various user requests. I can't do it.

Therefore, "variable sample value control" has been proposed in which the control cycle is varied in response to changes in dead time (for example, raw material flow rate). In this variable sample value control, the dead time τ _D resulting from the raw material flow rate is calculated based on the following equation (2), and then the dead time τ including the delay due to the time constant of the component analyzer as shown in the equation (3). _In this method, the control cycle τ is determined in consideration of _a and sampling control is executed.

[0009]

[Equation 1] τ = τ _D + τ _a (3) where F _i is the raw material flow rate.

[0010]

However, the variable sample value control as described above requires the dead time τ _{D of the} raw material flow rate.
Although the sampling time is determined in consideration of the change of, it is necessary to obtain the dead time τ _D from the integral equation (2), which is a very troublesome process. Further, in the case of component control of a distillation column or the like, one component analyzer is used to process many input points, but this causes a problem that the sampling time of the component analyzer is not constant. .

Further, as is apparent from the equation (1), the difference in deviation is multiplied by the proportional gain. Therefore, when the proportional gain is adjusted in advance to the disturbance, the reverse In addition, the proportional action will work excessively with respect to the change of the target value. Therefore, the value of the proportional gain for disturbance must be set to a value smaller than the appropriate value.

As a result, there is a problem that the proportional term does not sufficiently operate with respect to the disturbance, and slow hunting is repeated in a very long cycle. The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a control device that simply determines a control cycle from a sampling cycle of a component analyzer or the like and performs appropriate sampling control.

Another object of the present invention is to provide a control device which ensures proper operation by the proportional term and can exert high controllability even in a control system having a long dead time.

[0014]

In order to solve the above-mentioned problems, the invention according to claim 1 calculates using a predetermined control constant based on a deviation which is a difference between a target value and a controlled variable. And a control cycle calculating means for calculating the manipulated variable to be applied to the process, the control cycle calculating means determining the control cycle from the changing sampling cycle of the control quantity measuring device such as an analyzer, and the control cycle calculating means. The control device is provided with an integral term calculating means for performing an integral term calculation using a control cycle and obtaining an integral term calculation output.

Next, the invention according to claim 2 executes an operation using a predetermined control constant based on the deviation which is the difference between the target value and the control amount, and determines the operation amount to be applied to the process. In the control device for calculating, the integral term calculation means for obtaining an integral term calculation output using a predetermined control constant and a variable control cycle based on the deviation between the target value and the control amount, and the difference between the control amounts are used. The control device is provided with a proportional term calculating means for executing a proportional term calculation and an adding means for adding both calculated outputs of these calculating means to obtain the manipulated variable.

[0016]

Therefore, according to the invention corresponding to claim 1, by taking the above-mentioned means, the control cycle is increased by sequentially adding the predetermined processing cycles during the sampling cycle of the controlled variable measuring device such as the analyzer. Since it is determined and the integral term calculation is executed using this control cycle, it is possible to easily obtain the control cycle in consideration of the dead time due to the process and the dead time of the control amount measuring device. Sampling control can be executed.

Next, in the invention according to claim 2, the proportional term computing means multiplies the difference in the controlled variable by the proportional gain to extract the proportional term computing output, so that it is affected by the change in the target value. Even in a control system that operates properly against disturbance and has a very long dead time, hunting can be suppressed and good control can be executed.

[0018]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing the basic configuration of the device of the present invention. This control device includes a deviation calculation means 1 for obtaining a deviation e ′ which is a difference between a target value SV and a control amount PV, and a non-linear element 2 for performing a linearization process on the deviation e ′ obtained by the deviation calculation means 1. And a sample IP unit 3 for calculating the manipulated variable MV using the deviation e after the linearization processing by the non-linear element 2, the controlled variable PV and, for example, the selection signal s1 of one of a number of component analyzers. The controller 4 is constituted by the above, and the manipulated variable MV obtained here is applied to the process 5.

FIG. 2 is a diagram showing a detailed configuration of the sample IP section 3. The sample IP unit 3 detects a change in the component analyzer selection signal s1 shown in FIG. 3 and outputs an analyzer selection timing signal s2, and an analyzer selection timing from the change detection unit 31. When the signal s2 is received, as shown in s3 of FIG. 3, the previous control cycle τ _k-1
Is used, and the processing cycle CT in the control device 4 is sequentially added to obtain the current sampling cycle, that is, the control cycle τ _k , and the control cycle obtained by the control cycle calculating means 32. Integral term computing means 33 for computing an integral term for the deviation e _{k + 1} and the control amount PV
Proportional term calculating means 34 for calculating the proportional term using the difference of
And an adding means 35 for adding both operation outputs of these operating means 33 and 34.

The proportional term calculation means 34 subtracts the previous controlled variable from the current controlled variable and then multiplies the preset proportional gain K _P to obtain the calculated output of the proportional term.
Further, in the sample IP unit 3, when the selection timing signal s2 of the change detection unit 21 is ON, the output of the adding means 35 is output as the change amount ΔMV _n of the manipulated variable, and the selection timing signal s2 is other than ON. manipulated variable zero value in the case of △ MV other determination means 36 output as _n change in the previous operation amount output △ MV _n of the determination means 36 M
In addition to V _n-1 , a manipulated variable integrating means 37 for obtaining the manipulated variable MV this time is provided.

Next, the sample I constructed as described above
The operation of the P section 3 will be described based on the flow of FIG. First, when the operation starts, the number of analyzer selections k = 0,
An initialization process for setting the control period τ ₀ = T is performed (ST
1). This control cycle τ ₀ = T is a value determined in advance at the adjustment stage of the process, etc., and both k and τ ₀ are set in the control cycle calculating means 32.

Here, in the control period calculating means 32, the analyzer selection timing signal s output from the change detecting section 31 is outputted.
It is determined whether or not 2 is on "1", that is, whether or not to reset (ST2). Even if it is determined to reset, there is no control cycle to be initialized immediately after the initialization. Therefore, after the next step ST3, the increment process of k = k + 1 is performed.

Thereafter, a counting operation for sequentially adding the processing cycles CT of the control device is performed (ST4) to determine the control cycle τ _{k + 1} relating to the analyzer. At this time, the next new analyzer is selected. It is determined whether or not the timing signal s2 has arrived (ST5), and if it has not yet arrived, the counting operation is continued.

On the other hand, when the next new analyzer selection timing signal s2 arrives, the integral term calculating means 33
Using the deviation e _k this time, the control period τ _k-1 obtained by the counting operation, and the integration time T _I , I _k = (e _k · τ _k-1 ) / After performing the calculation of the integral term of T _I (4) (ST6), the control period is reset to τ _k-1 = 0 (ST7), and the S
Returning to T3, the same processing is repeatedly executed.

On the other hand, in the proportional term calculation means 34, the proportional calculation output is obtained in advance by multiplying the difference of the control amount PV by the proportional gain K _P , not by the deviation between the target value SV and the control amount PV. If the proportional gain K _P is set to an appropriate value with respect to the disturbance, the proportional action does not work excessively with respect to the change of the target value SV, and the proportional term appropriately operates with respect to the disturbance, and Even in a control system in which slow hunting is repeated in a very long cycle, hunting can be suppressed and good control can be executed.

In the above-mentioned embodiment, the description has been made on the assumption that the components of the distillation column are controlled. However, the present invention can be similarly applied to, for example, the control of the reaction vessel and various controls in which the dead time changes. In addition, the present invention can be modified in various ways without departing from the scope of the invention.

[0027]

As described above, according to the present invention, the following various effects are exhibited. According to the invention of claim 1, the control cycle can be easily determined from the sampling cycle of the analyzer and the like, and the sampling control can be appropriately performed. According to the second aspect of the invention, it is possible to secure an appropriate operation by the proportional term, and it is possible to exert high controllability even in a control system with a long dead time.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of a control device according to the present invention.

FIG. 2 is a detailed configuration diagram of a sample IP unit in FIG.

FIG. 3 is an operation explanatory diagram for obtaining a control cycle in FIG.

FIG. 4 is a flowchart illustrating the operation of the sample IP unit shown in FIG.

FIG. 5 is a diagram illustrating a conventional sample PI operation.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Deviation calculation means, 3 ... Sample IP part, 4 ... Control device, 5 ... Process, 31 ... Change detection part, 32 ... Control period calculation means, 33 ... Integral term calculation means, 34 ... Proportional term calculation means, 35 ... Addition means, 36 ... Judging means, 37 ... Operation amount integrating means.

Claims (2)

[Claims] 1. A control device for executing a calculation using a predetermined control constant based on a deviation, which is a difference between a target value and a controlled variable, to calculate a manipulated variable applied to a process, such as an analyzer. Control cycle calculating means for determining the control cycle from the changing sampling cycle of the control amount measuring device, and integral term calculation for calculating the integral term using the control cycle obtained by the control cycle calculating means to obtain the integral term calculation output. And a control device. 2. A control device for executing a calculation using a predetermined control constant based on a deviation, which is a difference between a target value and a control amount, and calculating an operation amount applied to a process, wherein: Integral term calculation means for obtaining an integral term calculation output using a predetermined control constant and variable control cycle based on the deviation from the control amount, and a proportional term calculation for executing a proportional term calculation using the difference between the control amounts. A control device comprising: a means and an addition means for adding both operation outputs of these operation means to calculate the operation amount.