JPH03268105A - Program controller - Google Patents

Abstract

PURPOSE:To preclude overshooting and undershooting and to perform control exactly according to a set pattern by performing forcible control to a next segment before a segment is switched and arithmetic control. CONSTITUTION:When a switching point discrimination means 104 discriminates that the forcible switching point of operation by a set value arithmetic means 101 reaches a point which is the idle time that a controlled system 103 has before the control end point of a segment under control, the switching means 102c of a manipulated variable arithmetic means 102 is brought under switching control in specific order. For example, when the next segment is a fixed value segment, the manipulated variable means 102 operates a manipulated variable by an I-PD arithmetic means 102a from a measured value from the controlled system 103 and a set value from the set value arithmetic means 101 and outputs the result to the controlled system 103. When the next segment is a lamp segment, on the other hand, the manipulated variable is operated by a PID arithmetic means 102 and outputted to the controlled system 103.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a program controller, and more particularly, to an improvement in a program controller that adjusts the temperature of, for example, an electric high temperature furnace or the like according to a preset program.

[Prior Art] Conventionally, as shown in FIG. 8, for example, a program controller of this type has a program pattern setting means 1, a set value calculation means 3, a comparison means 5, a PID calculation means 7, and a segment arrival determination means 9. , switching means 11, lamp constant storage means 1
3. It had a fixed value constant storage means 15 and a controlled object 17.

That is, the process section for controlling the controlled object 17 is divided into a plurality of sections as shown in FIG. 9, and each section is divided into segments (1) to
(5)...Binding, each segment 1-(1)-(5)...
- Segment identification corresponding to - - A control pattern is set by inputting segment data such as code, control time and control level from the program pattern setting means 1, and the set value calculation means 3 identifies segments corresponding to each segment according to the control pattern. The set value Sv is calculated and added to the comparison means 5, and the comparison means 5 obtains the deviation from the measured value PV from the controlled object 17 and adds it to the PID calculation means 7. 13
or PID based on the PID constant selectively added by the switching means 11 from the constant value constant storage means 15 and the above deviation.
It calculates and outputs the manipulated variable MV to the controlled object 17.

Further, when the segment arrival determining means 9 determines that the calculation of the set value SV has reached the calculation of the next control segment, the segment reaching determination means 9 switches and controls the switching means 11 to switch to the next control segment. It is configured to be able to output the corresponding PID constant to the PID calculation means 7.

Therefore, as shown in FIG. 10, in the double program controller, each segment is program operated in step 100, and it is determined in step 101 whether or not the next segment has been reached. Return to step 100 and repeat these steps 100 and 101.

If step 101 becomes YES, the following step 102
In step 103, the segment count is advanced by one, and in step 103, the PID constant is changed to a value corresponding to the advanced segment, and the process proceeds to step 100.

Therefore, for the ramp segment, a PID constant for follow-up control that has a fast response speed and quick response that follows the set value is stored in the ramp control constant storage means 13, and for the constant value segment, the PID constant for the disturbance control is stored in the ramp control constant storage means 13. The PID constant is stored in constant value control constant storage means 15, and segments (1) and (3
), (5).

, is the PI stored in the lamp control constant storage means 13.
The manipulated variable MV calculated using the D constant is segment (2),
(4), (6), , constant value control constant storage means 15
It is assumed that the manipulated variable MV calculated using the PID constant stored in is output to the controlled object 17, and the controlled object 17 is program-controlled according to the sequence of the entire segment, that is, the set control pattern, as shown in FIG. It is being

[Problems to be Solved by the Invention] However, in the program controller described above, it is desired that the controlled object 17 is program-controlled according to a preset control pattern, but if the controlled object 17 has a delay element or the like. Therefore, in reality, the following drawbacks tend to occur.

That is, after the calculation of the segment under control is completed, the PID constant corresponding to the next segment is determined by the PID calculation means 7.
Therefore, in the switching section from the ramp segment to the constant value segment, overshoot tends to occur at the control start section of the constant value segment, as shown in (a) in FIG. 9.

On the other hand, in the switching section from the constant value segment to the ramp segment, undershoot is likely to occur at the control start section of the ramp segment, as shown by (b) in FIG.

However, one set of PIDs can be used without switching the PID constants.
It is possible to operate the PID calculation means 7 using a constant value, but since the control response to a change in set value and the control response to a disturbance are different, if the optimum PID constant is selected for the ramp segment, that is, additional value control, the effect of the disturbance will be reduced in the constant value segment. On the other hand, if a PID constant that is optimal for the influence of disturbance is selected, the control response for the ramp segment, that is, the follow-up control, will not be optimal.

On the other hand, the 1-PD calculation method is well known as a control calculation method that uses one set of PID constants to equalize the response of the control characteristic to a change in a set value and the response of the control characteristic to a disturbance.

However, this method has poor ability to follow changes in set values, and when program control operation is performed, it becomes impossible to follow changes in set values from the start of lamp control, making it difficult to control according to the set program pattern. Not practical. The same applies when changing from fixed value control to ramp control.

The present invention has been made to solve these conventional drawbacks, and has a simple configuration that prevents overshoot and undershoot from occurring at the switching section from the currently controlled segment to the next segment. The purpose of the present invention is to provide a program controller that allows easy program control.

[Means for Solving the Problems] In order to solve such problems, the first configuration of the present invention is as shown in the claim correspondence diagram of FIG.
3. Program pattern setting means 10o1 is comprised of set value calculation means 101, operation amount calculation means 102, and switching point determination means 104.

The controlled object 103 is program-controlled according to the manipulated variable, and the program pattern setting means 100 defines each of a plurality of control process sections in which the controlled object is program-controlled as a segment, and sets at least the control level and the control level corresponding to each segment. The entire segment pattern is programmed by inputting the control time, and the forced switching point to the next segment before switching of this segment is set.

The set value calculation means 101 sequentially calculates the set amount from the control level and control time from the program pattern setting means 100.

The manipulated variable calculating means 102 includes an I-PD calculating means 102a that calculates the manipulated variable based on the measured amount from the controlled object 103 and the set amount from the set value calculating means 101, and an I-PD calculating means 102a that calculates the manipulated variable based on the measured amount from the controlled object 103 and the set value from the set value calculating means 101, and the I-PD calculating means 102a that calculates the manipulated variable based on the measured amount from the controlled object 103 and the set amount from the set value calculating means 101, and the I-PD calculating means 102a that calculates the manipulated variable based on the measured amount from the controlled object 103 and the set amount from the set value calculating means 101. -PD calculation means 102
P calculates the manipulated variable with a faster response speed to the set value than a.
It has an ID calculation means 102b and a switching means 102c that selectively switches between the I-PD calculation means 102a and the PID calculation means 102b.

The switching point determination means 104 controls switching of the switching means 102c of the manipulated variable calculation means 102 to the calculation means corresponding to the next segment when the calculation in the set value calculation means 101 reaches a forced switching point. .

In the first configuration, the forced switching point in the program pattern setting means 100 can be set before the end point of the 0 segment by about the dead time of the controlled object 103.

Moreover, as shown in the claim correspondence diagram of FIG. 2, the second configuration of the present invention has a controlled object 103 similar to the first configuration,
Program pattern setting means 100, set value calculation means 1
01, and a manipulated variable calculating means 105 and a switching point determining means 106 as described below.

That is, the manipulated variable calculation means 105 includes a PID calculation means 105a that performs PID calculation of the manipulated variable based on the measured amount from the controlled object 103 and the set value from the set value calculation means 101-, and PID calculations corresponding to the lamp and constant value segments. a plurality of PID constant storage means 105b for individually storing PID constants for lamp additional value control and fixed value control;
105c, and these PID constant storage means 105b, 1
05c to the PTD calculation means 105a.

The switching point determination means 106 includes a PID storage means 105b that stores a PrD constant corresponding to the next segment when the calculation in the set value calculation means 101 reaches a forced switching point;
105c is switched to the PID calculation means 105a.

Also in this second configuration, the forced switching point in the program pattern setting means 100 described in item 1 can be set before the end point of the segment by about the dead time of the controlled object 1.3.

[Function] In the first configuration of the present invention including such a means, the switching point determining means 104 serves as a forced switching point for calculation in the set value calculating means 101, and the switching point determining means 104 is used as a forced switching point for calculation in the set value calculating means 101. When it is determined that the object 103 has reached the front side by about the dead time, the switching means 102 of the operation amount calculating means 102
c in a predetermined order, and if the next segment is a constant value segment, the manipulated variable calculating means 102 performs
The PD calculating means 102a calculates a manipulated variable based on the measured value from the controlled object 103 and the set value from the set value calculating means 101, and outputs it to the controlled object 1.03.

1 2 On the other hand, if the next segment is a lamp segment, the manipulated variable calculating means 102 calculates the manipulated variable based on the measured amount and the set amount in the PID calculating means 102b, and outputs it to the controlled object 103.

Further, in the second configuration of the present invention, the switching point determination means 106
However, when it is determined that the calculation in the set value calculation means 101 has reached a forced switching point, the switching means 105d is controlled to switch in a predetermined order. A PID constant is input, and the PID calculating means 105a calculates the manipulated variable corresponding to the constant value segment based on the measured amount and the set amount, and outputs it to the controlled object 103.

On the other hand, if the next segment is a ramp segment, the operation amount calculation means 105 inputs a PID constant for lamp follow-up control, and the PID calculation means 105a performs an operation corresponding to the lamp segment based on the measured amount and set value. The amount is calculated and output to the controlled object 103.

[Example] Embodiments of the present invention will be described below with reference to the drawings.

FIG. 3 is a block diagram showing an embodiment of the program controller according to the present invention.

First, the first configuration will be explained.

In FIG. 3, the control device 19 includes a key input device 21,
It is connected to the storage device 23, output device 25, input device 27, and display device (not shown), and controls the human output of signals between them.
It is mainly formed of I9b, RAMI9c, and I1019d, and is the main part of a so-called microcomputer.

The CPU 19a operates the key input device 2 under a predetermined program.
1. In addition to controlling the input and output of signals between the storage device 23, the output device 25, and the input device 27, and calculating the set value SV and manipulated variable MV, the main functions described below are executed.

The ROM 19b stores programs that operate the CPU 19a, and the RAM 19c temporarily stores data such as the calculation process of the CPU 19a.
is an interface between the key input device 21-1, the memory 34 device 23, the output device 25, and the input device 27 and the control device 19.

The key input device 21 is, for example, a keyboard arranged on an operation panel (none of which is shown) of the main body of the program controller. Various segment data such as segment control time and final set value level for each segment can be entered.
A program pattern setting means 100 is formed to program a segment pattern as shown in FIG.

Further, the key input device 21 is under the control of the control device 19.
As shown in Figure 4, the calculation destination for a certain segment,
In other words, a predetermined period of time from the original calculation end point of the control period S of a certain segment, for example, before the dead time (1) of the controlled object 29 described later, or the manipulated variable calculation means 102 described later
Approximately 1/2 minute before the integral time calculated in is set as the forced switching point to the next segment.

The storage device 23 stores, under the control of the control device 19, each segment data inputted from the key input device 21, forced switching points, and control PID constants in correspondence with each segment, and is also readable. It was formed.

The output device 25 creates an analog operation iMV from the digital operation amount MV, which is the calculation result from the control device 19, and outputs it to the controlled object 29.
Depending on not only V but also the digital operation amount from the control device 19,
In some cases, digital signals are output at a predetermined rate within a certain period of time.

The controlled object 29 is, for example, an electric furnace, and is configured to output an analog measurement value PV detected by, for example, a built-in temperature sensor (not shown) to the input device 27 as feedback.

The input device 27 is under the control of the control device 19 to control the controlled object 2.
The analog measurement value P (feedback data) input from 9 is amplified and converted into a digital signal, and the signal is output as feedback to the control device 19.

=15 6 The control device 19 forms a set value calculating means 101 that reads the segment data from the storage device 23 in a predetermined order of segments according to the program stored in the ROMI 9 b, and creates a set value Sv. It takes in the measured value PV from the device 27, calculates the manipulated variable MV for each segment using the deviation between the set value SV and the measured value PV, and outputs it to the output device 25. is forming.

This operation amount calculation means 102, if explained with reference to FIG. I-1) D calculation means 102 that performs I-PD calculation based on the measured value PV from 103
a, and to compensate for poor tracking of set value changes, which is a drawback of the 1-PD calculation described above, during lamp control, these set values SV
and PID calculation means 102 that performs calculations based on the measured value PV.
b, and the manipulated variable MV is selectively output from the 1-PD calculation means 102a and the PID calculation means 102b.

As shown in FIG. 5, the calculation configuration of the I-PD calculation means and the PID calculation means includes, in addition to the switching means 102c, (1) an calculation circuit 102e and a P calculation circuit 102f.

Deviation differential PID calculation and I-P based on the connection configuration of the D calculation circuit 102g, comparison circuit 102d, and addition circuit 102h
A switching configuration for D calculation is known.

Furthermore, as shown in FIG.
Arithmetic circuit 102fXD arithmetic circuit 102g.

A measured value differential PID calculation and an I-PD calculation configuration using a connection configuration of the comparison circuit 102d and the addition circuit 102h are known, but since both are well known, their explanation will be omitted.

In FIG. 3, the control device 19 determines whether or not the computation of the set value Sv in a certain segment has reached the forced switching point set by the key input device 21, and when it is determined that the compulsory switching point has been reached, the control device 19 as shown in FIG. and switching means 10 in FIG.
2c is also formed.

Therefore, the control device 19 performs a PID calculation that has relatively good followability for setting changes during lamp control, for example, when the lamp control time = 17 8, and when the calculation of the set value S■ in the lamp segment reaches the forced switching point. I” - Switches to PD calculation, and then calculates the operation ftMV with the switched calculation configuration and outputs the program, and sets the segment count to 1.
When the compulsory switching point is reached during the computation of the set value Sv in the next segment, the computation is switched to PID computation.

Note that when switching the calculation configuration, bumpless operation is performed to avoid fluctuations in the manipulated variable MV.

The program controller of the present invention configured as described above will become clearer from the flowchart shown in FIG.

When the program starts, program operation is executed in step 700, and the digital operation amount MV of the programmed segment is calculated and output to the output device 25.
For example, the analog manipulated variable MV is outputted from the output device 25 to the controlled object 29 .

In the following step 701, it is determined whether the next segment has been reached. If NO, the process moves to step 702, and it is determined whether the forced switching point has been reached in the segment under control. If this step 702 is NO, The program returns to step 700, and the program operation is repeated until step 700 or 702 becomes YES.

If step 702 becomes YES, it is determined in step 703 whether the operation mode of the next segment is a fixed value, and YES.
In the case of S, that is, when the next segment is a constant value segment, the control calculation is bumplessly switched to the I-PD calculation in step 704, and the process returns to step 700.

If step 703 is No, that is, if the next segment is a ramp segment, control calculation is bumplessly switched to PID calculation in step 706, and step 70
Return to 0.

Then, PID calculation or I-PD according to step 700
When step 701 determines the arrival of the next segment during program operation based on calculation, step 701 returns YE.
S, and the process moves to step 706.

=19-=20 In step 706, the segment counter is incremented by one and the process returns to step 700.

Therefore, in the program controller of the present invention, the calculation of the set value S is forced during program control (more precisely, set value calculation) of the controlled object 29 according to the program pattern input from the key input device 21 in FIG. When the switching point is reached, the control calculation is automatically switched to the one corresponding to the next segment, so even if there is a delay time in the controlled object 29, overshoot or undershoot will not occur even when moving from the ramp segment to the constant value segment. This makes overshoot and undershoot less likely to occur even when moving from a constant value segment to a ramp segment.

Therefore, the controlled object 29 is program-controlled according to the program pattern.

Next, a second configuration of the present invention will be explained with reference to FIG.

In the first configuration described above, the operation amount MV calculation means is I-P.
It is composed of a D calculation means 102a and a PID calculation means 102b, and is configured to be switched by a switching means 102C, and the algorithm is changed by changing the algorithm while keeping the PID constant constant.In the present invention, however, the operation amount calculation means 1
05 may be configured as follows.

That is, the PID calculation means 105a and the lamp control P
Lamp control PID constant storage means 1 storing ID constants
05b, constant value control PID constant storage means 105C that stores constant value control PTD constants, and switching means 105d that switches these storage means and selectively outputs the PID constant to the PID calculation means 105a. When the switching point determining means 106, which is similar to the switching point determining means 104 described above, determines that the forced switching point has been reached, the switching means 105d is controlled to switch.

In addition, these lamp control P■D constant storage means 105b
The constant value control PID constant storage means 105c is formed by the storage device 23 shown in FIG. 3, and the PID calculation means 105a and the switching means 105d are formed by the control device 19 shown in the same figure.

In this second configuration, the calculation algorithm remains unchanged and the PID constants are switched in correspondence with the ramp and constant value segments, and the processing in steps 704 and 705 in FIG. 7 is performed using the PID constants for constant value control. P.I.D.
Other processing steps are the same as in the first configuration except for the calculation processing or the PID calculation processing using the lamp PID constant.

In the present invention, the forced switching point from the segment under control to the next segment can be set arbitrarily, and as shown in FIG. ) minute before, a certain degree of effect can be expected, but in consideration of the controlled object 29, it is preferable to set the time point to about (1) minute before the dead time of the controlled object 29.

In addition, recent programmable controllers automatically test control the controlled object 29 to automatically set the optimum PID constant for the controlled object 29 and perform PID calculation.
Program controllers with a built-in so-called PID constant auto-tuning (self-tuning) function are available, but even in programmable controllers with this built-in auto-tuning function, the dead time is determined from the automatically set PTD constant. Switching points can be set automatically.

[Effects of the Invention] As explained above, the present invention is capable of preventing overshoot or undershoot from occurring at the switching portion from one segment to the next segment in program pattern control.
Good followability to set value changes during lamp control,
During constant value control, an optimal control response to the influence of disturbances is possible, program control according to a set pattern is facilitated, and the configuration is less complicated than that of a conventional program controller.

In particular, in the first configuration of the present invention, there is no switching of PID constants and one set of PID constants is used, which has the advantage of simplifying the constant setting operation.

[Brief explanation of drawings]

FIG. 1 is a claim correspondence diagram relating to the first configuration of the present invention, FIG. 2 is a claim correspondence diagram relating to the second configuration of the present invention, and FIG.
The figure is a block diagram showing an embodiment of the program controller 34 according to the present invention, FIG. 4 is a diagram showing a control program pattern explaining the operation of the program controller of FIG. 2, and FIGS. A block diagram showing the PID calculation configuration in the present invention, FIG. 7 is a flowchart explaining the operation of the program controller of the present invention, FIG. 8 is a block diagram showing a conventional program controller, and FIG. 9 is a conventional control program. FIG. 10, which is a diagram showing the pattern, is a diagram explaining the operation of FIG. 8. 1...Program pattern setting means 3...
...Setting value calculation means 5 ...Comparison means 7 ...PiD calculation means 9 ...Segment attainment determination means 11 ...
...Switching means 13...Lamp control constant storage means 15...
...Constant storage means for constant value control 1-7.29...
・Controlled object 19...Control device 21...Key input device 23...・・・ 101 ・ ・ ・ ・ ・ 102, 105 ・ 102 a ・ ・ ・ 102b ・ ・ ・ 102c, 105 103 ・ ・ ・ ・ 104, 106 ・ 105 a ・ ・ −・ 105 b ・ ・ ・ ・ Memory Device, output device, input device, program pattern setting means, setting value calculation means, operation amount calculation means, I-PD calculation means, PID calculation means d...switching means, controlled object, switching point discrimination means, PID calculation means lamp Control PID constant storage means 105c ・ ・ ・ ・ Fixed value control PID constant storage means 105 d...Switching means

Claims (4)

[Claims] (1) A control object that is program-controlled by a manipulated variable and each of a plurality of control process sections that programmatically control this control object are defined as segments, and at least the control level and control time corresponding to each segment are input to determine the overall program pattern setting means for programmatically setting a segment pattern and setting a forced switching point to the next segment before switching the segment; and sequentially calculating set amounts from the control level and control time from the program pattern setting means. a set value calculating means for calculating the manipulated variable based on a measured amount from the controlled object and a set amount from the set value calculating means; PID calculation means for calculating the manipulated variable having a faster response speed to the set value than the I-PD calculation means; and switching means for selectively switching between the I-PD calculation means and the PID calculation means. and switching point determining means for controlling the switching means of the manipulated variable calculating means to switch to the calculating means corresponding to the next segment when the calculation in the set value calculating means reaches the forced switching point. A program controller comprising: and . (2) The program controller according to claim 1, wherein the forced switching point in the program pattern setting means is approximately a dead time of the controlled object from the end point of the segment. (3) A control object that is program-controlled by the manipulated variable and each of a plurality of control process sections that program-control this control object are defined as segments, and at least the control level and control time corresponding to each segment are input to determine the overall program pattern setting means for programmatically setting a segment pattern and setting a forced switching point to the next segment before switching the segment; and sequentially calculating set amounts from the control level and control time from the program pattern setting means. a set value calculation means for calculating the manipulated variable based on a measured amount from the controlled object and a setting amount from the set value calculation means, each having a different response speed to the set value, and a lamp and a A plurality of PID constant storage means for individually storing PID constants for lamp control and constant value control used in PID calculation corresponding to the constant value segment, and these PID constant storage means are selectively connected to the PID calculation means. switching means; when the calculation in the set value calculation means reaches the forced switching point, the PID storage means storing the PID constant corresponding to the next segment is switched to the PID calculation means; A program controller characterized by comprising: a switching point determining means for controlling. (4) The program controller according to claim 3, wherein the forced switching point in the program pattern setting means is approximately a dead time of the controlled object from the end point of the segment.