JP2855644B2 - Controller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a PID controller which obtains a control (operation) output by performing a proportional (P), an integral (I), and a differential (D) operation. More particularly, the present invention relates to a controller that corrects a set value to be a control target based on a correlation between a set target value and a process amount so that the process amount can be settled to the set target value smoothly.

<Prior Art> FIG. 8 is a conceptual diagram showing a configuration of a conventional example of a PID controller. In the figure, 1 is a process, 2 is a process amount PV from the process 1 and a set value SP, and a control output to the process 1 is provided.
The PID calculator 3 for giving the MV is a rate limiter to which the set target value TSP is applied and outputs the set value SP to be given to the PID calculator 2.

When the rate limiter 3 is turned on, the PID calculator 2
Is a signal that changes in a ramp shape due to the change rate limitation here, and a signal that changes in a step shape when turned off.

FIG. 9 is a diagram showing a general response of the control output MV and the process amount PV when the set value SP is changed in a ramp shape in FIG.

In response to the change of the set value SP, the control output MV changes as shown in the figure by PID calculation, and as a result, the process amount PV changes so as to follow the set value SP with a slight delay.

<Problem to be Solved by the Invention> However, the conventional apparatus has a problem that the process amount PV slightly overshoots even after the set value SV is maintained at a constant value, and the responsiveness is impaired.

The present invention has been made in view of such problems, and a controller capable of preventing overshoot at the time of settling of the process amount PV and performing stable control without impairing responsiveness at startup. The purpose is to provide.

<Means for Solving the Problem> FIG. 1 is a principle block diagram of the present invention.

In the figure, 1 is a process, 2 is a PID calculation unit that inputs a process amount PV from the process 1 and a set value SP1 and performs a PID calculation, and provides a control output MV to the process 1; The applied rate limiter.

Reference numeral 4 denotes a set value correction calculating unit which inputs the set target value TSP and the process amount PV from the process, and sets the set target value TS.
The difference EV between P and the process amount PV and the change amount DPV of the process amount
The set target value is corrected based on the correlation with, and the corrected set value SP1 is output to the PID calculation unit.

<Operation> The set value correction calculation unit 4 applies a fuzzy rule based on a difference EV between the set target value TSP and the process amount PV and a change amount DPV of the process amount PV to obtain a correction amount of the set value SP. The corrected set value is given to the PID calculation unit 2.

As a result, it is possible to finely correct the set value with respect to the input of various conditions, thereby realizing stable control.

<Example> Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention. Here, the internal configuration of the set value correction calculating section 4 in FIG. 1 is shown.

Reference numeral 41 denotes a margin calculation unit (adder) for inputting the set target value TSP and the process amount PV from the process 1, and calculating the difference between the set target value TSP and the process amount PV, that is, the reach margin EV. Reference numeral 42 denotes a PV change amount calculating means for inputting a process amount PV and calculating a PV change amount DPV per fixed time. Here, the certain time is, for example, the equivalent dead time of the process 1 and is a measure of the overshoot amount of the process.

Reference numeral 43 denotes a fuzzy rule application unit which inputs the margin EV and the PV change amount DPV from the margin calculation unit 41, and outputs a correction amount SPC based on a correlation between these amounts.

44 is an adder for adding the set attainment target value TSP and the correction amount SPC, 45 is an addition signal TSP + SPC from the adder 44 and the set value SP
The switch means 46 selects and outputs one of them, and the input means 46 receives the addition signal TSP + SPC, the set target value TSP, and the set value SP, and drives the switch means 45 according to these values.

The judging means 46 compares the addition signal TSP + SPC with the set value SP, and when SP <TSP + SPC, connects to the b side to select the set value SP and outputs the set value SP to the PID calculation unit 2, where SP ≧ TSP + SPC At the time, it is configured to connect to the a side to select the addition signal TSP + SPC, output the selected signal to the PID calculation unit 2, and hold until the set target value TSP changes next.

FIG. 3 is a diagram showing fuzzy variables used in the fuzzy rule application unit 43.

The horizontal axis is the input variable, which is normalized to [−1, 1].
The vertical axis represents the output, and five fuzzy variables as shown.
NB, NM, ZO, PM, PB are set. Here, the variable NB is negative and large, NM is negative and ordinary, ZO is zero, PM is positive and ordinary, and PB is positive and large.

The two inputs DPV and EV to the fuzzy rule application unit 43 are
Each is normalized by multiplying by 1/100 (1 /%), and is assigned to five fuzzy variables by correlation of these values.

FIG. 4 is a diagram showing an example of a combination of rules to be assigned to five fuzzy variables from the correlation between two values.

These rules are based on "if DPV is PB and EV is ZO TH
EN SPC is NB ", that is," If DPV is positive large and EV is zero, make SPC negative large ".

In estimating the correction amount SPC from these rules, the centroid method is used here.

 FIG. 5 is an explanatory diagram of this centroid method.

 In the figure, the inference method is "if x is X1 and if y is Y1 then z is Z1" and "if x is X2 and if y is Y2 then z is Z2".

Since the if part (referred to as the antecedent part) is "and", it means that in the fuzzy operation, "and" takes a Min value, which takes a value as shown by a horizontal dotted line in FIG.
After then (referred to as the consequent part), the output of the antecedent part is the area of the lower part obtained by cutting the fuzzy variables, which is
Corresponds to the shaded portions of Z1 and Z2.

The output synthesis takes the logical sum (OR) of the areas, finds the center of gravity, and outputs the value of z corresponding to the center of gravity.

The output fuzzy variables are the same as those in FIG. 3, and the actual correction amount SPC is obtained by multiplying the input variable × 100%.

FIG. 6 is a diagram showing an example of a simulation in the case where control is performed by applying a fuzzy rule in the apparatus of the present invention. The operation will be described below with reference to FIG.

In this example, a step response is shown so that the movement of the corrected set value SP1 given to the PID calculation unit 2 can be easily understood. In the step response, set value SP = set target value TSP
Therefore, when the correction amount SPC> 0, the corrected set value SP1 is output.

First, at time to, the set attainment target value TSP is given in steps (70%). In response to this, the PID calculation unit 2 sends 1
00% MV output is output to process 1. Until the time t ₁ is, "EV (allowance) is PB" or "EV is PM" (4th
In this case, “EV is PM” and “DPV
The correction amount SPC is “0” unless “PB” is obtained. Near this time, the process amount PV is accelerated and increasing, and the margin EV gradually decreases.

I was past the time t _1, ie, as the margin amount EV is reduced, "EV is ZO", "EV is PM" or, "DPV is P
The conditions of “M” and “DPV is ZO” are satisfied, and the correction amount SPC is output according to the SPC weight determined by the combination.
As a result, set value SP1 given to the PID calculating unit 2 is motion corrected to slightly decrease from 70% of the values as shown from time t _1.

When the process amount PV further approaches the set target value TSP and the PV change amount DPV approaches “0”, “DPV is Z
The effect of “O” becomes strong, and the correction amount SPC also decreases and converges.

Thus, the process amount P with respect to the set target value TSP
When the margin (deviation) EV of V is large, the correction amount SPC is “0”, but when the EV is medium or small, it is set according to the DV change amount (expected process overshoot amount) DPV. The value SP is corrected to be small, and braking is applied.

Further, for example, when the process amount PV moves in the reverse direction (direction in which the deviation increases) and the margin EV is increasing, it corresponds to a blank portion in FIG. 4 and does nothing here. I have.

By such an operation, the set value SP1 given to the PID calculation unit 2 is corrected with time as shown in FIG.
The process amount PV smoothly reaches the set target value TSP without overshoot.

FIG. 7 is a diagram showing an example of a control simulation when the set value SP is changed from ramp to soak. In this case, the same processing is performed for the set target value TSP, and when SP> TSP + SPC, by changing from SP to TSP + SPC, both following the ramp and preventing overshoot are achieved. I have.

In the above embodiment, the fuzzy rule application unit 43
Although the example in which the margin EV and the change amount DPV of the process amount are input has been described, for example, the change rate of the change amount DPV, the set target value TSP, or the like may be given. Control becomes possible. In addition, a calculation method other than the centroid method may be used. Furthermore, in applying the fuzzy rule, a fuzzy operation may be performed each time, and a part of the operation may be omitted by preparing a table in advance to improve the operation speed as a whole. It may be.

<Effects of the Invention> As described in detail above, according to the present invention, the set value given to the PID calculation unit is set to at least a margin EV (deviation),
Correction is performed based on the PV change amount DPV, and stable control can be performed by preventing overshoot without losing adaptability. Further, various conditions can be input by adopting the fuzzy method as in the embodiment, the set value can be corrected more human-like, the change can be smooth, and the change in the operation amount MV can be prevented from bumping.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the principle of the present invention, FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 3 is a diagram showing fuzzy variables used in a fuzzy rule application unit,
FIG. 4 is a diagram showing an example of a combination of rules assigned to five fuzzy variables from the correlation between two values, FIG. 5 is an explanatory diagram of a centroid method, and FIG. FIG. 7 is a diagram showing an example of a simulation in which control is performed by using a control method, FIG. 7 is a diagram showing an example of a control simulation in which a set value is changed from ramp to soak, and FIG. FIG. 9 is FIG.
FIG. 7 is a diagram showing a general response of a control output and a process amount when a set value is changed in a ramp shape in the figure. 1 Process 2 PID calculation unit 3 Rate limiter 4 Set value correction calculation unit 41 Margin calculation unit 42 PV change calculation unit 43 Fuzzy rule application unit 44 Adder, 45: Switch means 46: Judgment means

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-63-55602 (JP, A) JP-A-64-3703 (JP, A) Jikai Sho 62-129603 (JP, U) (58) Field (Int.Cl. ⁶ , DB name) G05B 11/36 505 G05B 13/00-13/02 JICST file (JOIS)

Claims (2)

(57) [Claims] 1. A process amount is input from a process (1) and PI
P that performs D operation and gives control output (MV) to the process
An ID calculation unit (2), a rate limiter (3) to which a set target value (TSP) is applied, and the set target value and a process amount (P
V), the difference (EV) between the set target value (TSP) and the process amount (PV) and the change amount (DPV) of the process amount
And a set value correction calculating section (4) for correcting the set target value from the correlation with the set value and outputting the corrected set value (SP1) to the PID calculating section. 2. The set value correction calculator receives a set value (SP) from a rate limiter in addition to the input, and calculates a difference (EV) between the set target value (TSP) and a process amount (PV). ) And a means (43) for obtaining a correction value (SPC) of the set target value from a correlation between the process amount change amount (DPV), and a correction value of the set target value and the set target value. Switch means (45) for selecting one of the added value (SPC + TSP) and the set value, and determining means (46) for comparing the added value with the set value and selecting the smaller one. 2. The method according to claim 1, further comprising: selecting a value selected by the determination unit by the switch unit, and outputting the selected value as the corrected set value (SP1) to the PID calculation unit. Controller.