JPH0264801A - Process controller system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to an improvement of a process controller system with a backup function.

(Prior Art) FIG. 3 is a block diagram schematically showing the configuration of a conventional process roller system.

In the figure, reference numeral 43 denotes a mask controller 1n, which normally operates according to different control algorithms defined by the user.

Each controller 10 receives a signal from the process input device 9n and executes control calculations such as PID calculations.
Manipulate operational output.

7n is a conversion unit that outputs the operation output to the operation output device 8n, and each controller 1n is connected to this conversion unit 7n.
Each conversion unit 7n is controlled so that the output becomes the operation output value calculated by the control m calculation.

The conversion unit 7n includes a counter that goes up and down under control from the controller, a D/A converter that converts the contents of the counter into voltage, a V,/1 circuit that converts the voltage into current, and a D/A converter. A circuit for returning the output of the converter to the controller side is included.

This unit 7n is designed to have extremely high reliability, so there are actually only two parts where failure can occur.
It is now limited to the controller side.

As a result, even in a single system without a backup function, or in a redundant system with a backup function as shown in FIG. 3, at worst five units 70 can hold the operating output.

Also, when switching to backup, unit l-7
Since n is switched while holding the manipulated output value, there is also the advantage that bumps do not occur during switching.

Reference numeral 2 denotes a slave controller, one of which is prepared for a plurality of mask controllers 1n.

A switch 4n is provided in pairs with each mask controller so that the input/output of the slave controller 2 can be switched with the input/output of any mask controller 10.

Inside each switch 4n, an input signal switching means 6n and an operation output control signal switching means 5n are provided.

Reference numeral 3 denotes a backup memory controller, which constantly collects control algorithms, control parameters, tracking data necessary for continuation of control, etc. of the mask controller 1n through the transmission line 10.

The backup memory controller 3 stores the mask controller 1 due to a transmission error with the mask controller 1n or due to self-diagnosis information of the mask controller 1n itself.
detects an abnormality in the mask controller 1n, transfers various information of the corresponding mask controller to the slave controller 2, operates the switching device 4n paired with the corresponding mask controller 1n using the switching control line 11, and switches the slave controller 1 to the roller 2. Switch the input/output lines and prepare for backup.

After the backup preparations are completed, the slave controller 2 replaces the master controller [1-ra 1n (enters backup operation).

In this way, traditional process controller systems have a
3, a plurality of digital regular process controllers (equivalent to the master controller 10) and one digital backup process controller (equivalent to the slave 1 controller 2) to take over the functions of any of them. and a backup memory controller 3 that manages control algorithms, control parameters, tracking information necessary for continuation of control, etc. of each of the regular process controllers, and are connected by a transmission line 10.

Then, the backup memory controller (corresponding to the slave controller 2) sequentially and cyclically collects tracking information etc. from each regularly used process controller (corresponding to the master controller 1n) through the transmission line 10, and at the same time collects tracking information etc. from each commonly used process controller (corresponds to the master controller 1n). When a failure is detected in a process controller (equivalent to master controller 1n), the control algorithm, control parameters, and tracking information of the regular process controller (equivalent to master controller 1n) are transferred to the backup process controller (equivalent to slave controller 2). The backup process controller is set to take over the functions of the regular process controller.

Further, each of the regular and backup process controllers uses one that finally outputs a manipulated output value calculated from the deviation between the process input value and the set value as a pulse train signal.

For this reason, a configuration is adopted in which the pulse train signal is converted into a current signal in the unit 7n as a current signal for finally operating the operation output device 8n.

(Problem to be Solved by the Invention) However, in such a conventional process controller system, each control loop is configured separately into the mask controller 1n and the conversion unit 7n, resulting in a poor space factor and the conversion unit 7n. Since the mask controller 1n is designed to be highly reliable, the price of the unit 7n becomes an advantage, and the signal output from the mask controller 1n takes the form of a pulse train, which causes problems such as being a source of noise.

Therefore, in order to solve this problem, it is conceivable to incorporate the function of the conversion unit 7n into the mask controller 1n. However, if such a configuration is adopted, bumps may occur due to discrepancies in tracking information during backup switching. There is a problem that this occurs.

However, the backup memory controller 3 sequentially and cyclically collects tracking information and the like from each mask controller 1n through the transmission line 10.
Although the control calculation cycle for each mask controller 1n(,:d3) becomes longer as the number of connected units increases, the control calculation cycle for each mask controller 1n(,:d3 is considerably shorter than this. Therefore, if tracking information is collected via the transmission line 10, Then, if a failure occurs in any of the mask controllers 1n,
Contents of tracking information stored in the backup controller 3 and mask controller 1 related to the failure
A discrepancy occurs between the data contents of n and the result is a so-called bump in the operation output at the time of backup switching.

This invention was made in view of the above-mentioned problems, and its purpose is to reduce costs by eliminating the conversion unit in this type of process controller system, and to prevent bumps from occurring during backup switching. There is a particular thing.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above object, the present invention provides a plurality of digital regular process controllers and a third one that takes over the functions of any of them. A digital backup processor controller and a backup memory controller that manages control algorithms, control parameters, tracking information necessary for continuation of control, etc. of each of the regular process controllers are connected to each other by a transmission path, and the backup memory The controller sequentially and cyclically collects tracking information etc. from each regular process controller through the transmission path, and if a failure is detected in any regular process controller during the collection, the control algorithm and control of the regular process controller are In the process controller system, parameters, tracking information, etc. are set in the backup process controller so that the backup process controller takes over the functions of the regular process controller.

As each of the regular and backup process controllers, one that finally outputs the operation output value calculated from the deviation between the process input value and the set value as a current value is used, and each of the regular process controllers In this method, a switching device is interposed to switch the process input device and the manipulated output device from the regular process controller side to the backup process controller side, and directly transfer the manipulated output value of the regular process controller to the backup process controller, When the backup process controller takes over for a failed machine, it sets the manipulated output value directly transferred from the regular process controller as its own initial manipulated output value by operating the switching device, and then starts the backup operation. It is characterized by:

(Function) According to this configuration, a controller is used as each regular and backup process controller that outputs the operation output value calculated by calculation from the deviation between the process input value and the set value as a power value. This eliminates the need for the so-called conversion unit in conventional systems, reducing costs.Also, when a backup process controller takes over for a failed machine, the transfer is performed directly from the regular process controller by operating a switch. Since the backup operation is started after the input operation output value is set as the own operation output initial value, so-called mismatches in tracking information are eliminated, and bumps do not occur during backup.

(Embodiment) FIG. 1 is a block diagram schematically showing the entire process controller system according to the present invention.

As shown in the figure, in the system of the present invention, the mask controller 1n and the slave controller 2 finally output the manipulated output value calculated from the deviation between the process input value and the set value as a current value. is used.

Therefore, the conversion unit 7n shown in FIG. 3 becomes unnecessary, and the cost is reduced accordingly.

In addition, since the output values of the mask controller 1n and the slave controller 2 are current values, when backing up either of the mask controllers 1n with the slave controller 2, it is necessary to
If the output value of the slave controller 2 is not switched to match the output value of the slave controller 2, bumps will occur, the operating end will fluctuate, and the control system will be adversely affected.

Therefore, in each switch 4n, the process input device 9n and the manipulated output device 8n are switched from the mask controller 1n side to the slave controller 2 side (corresponding to switching means 5n and 6n), and the manipulated output value of the mask controller 1n is A new built-in function (not shown in FIG. 1) is to directly transfer the data to the slave controller 2.

FIG. 2 is a block diagram showing in detail the internal configuration of the mask controller 1n1, the slave controller 2, and the switch 4n.

As shown in the figure, the mask controller 1n includes a microprocessor 100 that executes control calculations, a transmission interface 101 that controls transmission, and an A/D
The converter 102, the D/A converter 103, the V/1 circuit 104 that converts voltage into current, AD converts the output of the own D/A converter 103, or the slave controller 2
A switching means 105 for selecting whether to perform AD conversion on the output device of the D/A converter, connected to the current output and output port 10;
A transistor 106 whose on/off state can be controlled by the output of the microprocessor 100, and a ROM, RAM (not shown), etc. necessary for the operation of the microprocessor 100 are built in.

Similarly, microprocessor 1003. Transmission interface 101s
, A/D converter 102s, D/A converter 1033. V/
I circuit 10431 switching means 105S1 transistor 10
6s, ROM, RAM (not shown), etc. are built-in.

In addition, the output boat 107.107s has two other types of output, which are not used when functioning as a mask controller, but are used to operate the switch 4n when functioning as a slave controller. Ru.

On the other hand, the switch 4n is constructed as follows.

That is, the transistor 200 is connected to the mask controller 1
It is connected to the current output from n, and is controlled on and off by opening the AND gate 201 based on the slot designation of the slave controller 2 and the output 302 of the slave controller 2.

Here, slot designation is performed by a signal line 300,
This signal is usually 3-bit information composed of 3 trees,
In other words, it is possible to specify 8 slots (23=8).

The slot designation from the 7C1 slave controller 2 is confirmed by the slot match determination circuit 203.

The switching means 204 is for connecting the output of the D/A converter 103 of the mask controller 1n paired with each switch 4n to the A/D converter 102s of the slave controller 2.

The switching means 205 is for connecting the current output of the slave controller 2 to the operation output device 8n.

These switching means 204 and 205 perform an AND operation between the slot designation of the slave controller 2 and the slave output 301.
On/off control is performed by opening the gate 202.

Further, the AND gates 206 and 207 are diodes, and are used to prevent reverse flow of current.

Next, the operation of the process controller having the above configuration will be sequentially and systematically explained with reference to FIG.

In the mask controller 1n, normally, by connecting the switching means 105 to the output of the self-D/A converter 103,
Self-diagnosis is performed to see if the manipulated output is being converted correctly to D/A.

Further, the transistor 106 is always off, and the current output is output to the switch 4n.

The slave controller 2 normally connects the switching means 105s to the output of its own D/A converter 103s, similar to the mask, to give a test signal to the D/A converter and check whether the D/A converter is correctly converted. I am self-diagnosing.

On the other hand, the transistor 106s is turned on to prevent its current output from flowing out.

Normally, the switch 4n sends the current output from the mask controller 1n to the operation output device 8n, with the transistor 200 turned off and the switching means 204 and 205 both turned off.

When an abnormality occurs in the mask controller 1n, the slave controller 2 receives various information about the corresponding mask controller from the backup memory controller 3.

The slave controller 2 is the corresponding mask controller 1
In addition to specifying the slot of the switch 4n that is paired with n, the control line 301 is activated to switch the switching means 204.
．． Turn on 2.05.

By turning on the switching means 204, the D/A converter output of the faulty mask controller is input to the slave controller 2 through the signal [1303, so the slave controller 2 is connected to the signal line 303 by the switching means 105S, and the D/A converter output of the faulty mask controller is input to the slave controller 2 through the switching means 105S. /A converter output value directly
/D conversion and reading.

The slave controller 2 sets the value read directly without going through the transmission line 10 to its own D/A converter 1033, and then activates the control line 302 to turn off its own transistor 106S and switch. The internal transistor 200 is turned on.

Therefore, the current output of the slave controller 2 flows out toward the switch 4n because the transistor 106S is turned off, and the current output of the mask controller is consumed within the switch when the transistor 200 is turned on.
The current from the slave controller 2 begins to flow to the operation output device 8n, completing the back-up amplification.

Next, the operation when canceling backup will be explained.

When the failed mask controller is restored by replacement or the like, the switching means 105 is connected to the signal line 304, which causes the D/A converter output of the slave controller 2 to become A/A.
It is converted into D and set in its own D/A converter 103.

The backup cancellation command is generally performed by a person who confirms that the faulty controller has returned to normal and then restores the original state based on human instructions.The CRT connected to this system
It is delivered to the backup memory controller 3 through a transmission line 10 from a device or the like (not shown).

The backup memory controller 3 tracks its own output with the output of the slave controller 2 so that the backup can be canceled at any time immediately after recovery, so the slave controller 2 connects the control line 302 to the output of the slave controller 2.
is inactivated to turn on its own transistor 106S and turn off the switch transistor 200.

When this occurs, the slave controller 2 makes the control line 301 inactive, turns off the switching means 204 and 205, and cancels the backup state.

In this way, in the system J3 of this embodiment, if any of the mask controllers 1n fails, the switching means 204 is turned on in the corresponding switching device 40, and the output of the D/A converter 103 is switched to the slave state. controller 2
and then initialized to the D/A converter 103S.

At the same time, the switching transistor 200 is turned off and the switching means 205 is turned on, and the process input/output related to the failure is automatically switched from the mask controller 1n side to the slave controller 2 side.

As a result, during backup switching, the final output value of the mask controller is always held while switching to the slave controller side, and no bumps occur.

That is, if the output initialization at a3 of the slave controller 2 is performed based on the tracking information from the transmission line 10, a bump will occur at the time of backup switching due to the difference between the transmission cycle of the tracking information and the calculation cycle of the mask controller. On the other hand, in this embodiment, the output value of the faulty mask controller is directly read through the control line 303 instead of being based on tracking information, so there is a difference between the tracking information transmission cycle and the mask controller calculation cycle. Even if there is, no bumps will occur during backup switching.

In the above embodiment, the switching operation in each switch 4n was performed by a command from the slave controller 2, but as in the conventional example shown in FIG. Of course, you may do so.

[Effects of the Invention] As is clear from the above explanation, according to the present invention, the expensive conversion unit required in the conventional system is no longer necessary, and the cost can be reduced by that amount. It has the advantage that it does not occur.

[Brief explanation of the drawing]

FIG. 1 is a block diagram schematically showing the overall configuration of the system of the present invention, and FIG. 2 is a block diagram showing details of the system.
FIG. 3 is a block diagram schematically showing the overall configuration of a conventional system. 1n...Mask controller 2...Slave controller 3...Backup memory controller 4n...Switcher 5n...Switching means 6n...Switching means 8n...Manipulation output device 9n...Process input device 10...transmission line 200...switcher transistor 204...switching means 205...switching means

Claims (1)

[Claims] A plurality of digital regular process controllers, one digital backup process controller for taking over the functions of any of them, and control algorithms and controls for each of the regular process controllers. A backup memory controller that manages parameters, tracking information necessary for continuation of control, etc. is connected to each other by a transmission line, and the backup memory controller sequentially and cyclically collects tracking information, etc. from each regular process controller through the transmission line. At the same time, when a failure is detected in any of the regularly used process controllers during the collection, the control algorithm, control parameters, tracking information, etc. of the regularly used process controller are set in the backup process controller and used regularly in the backup process controller. In a process controller system that takes over the functions of a process controller, each of the regular and backup process controllers uses the manipulated output value calculated from the deviation between the process input value and the set value as the final current. In addition, for each of the regular process controllers, the process input device and the manipulated output device are switched from the regular process controller side to the backup process controller side, and the manipulated output value of the regular process controller is used. When a switching device is used to directly transfer data to a backup process controller, and when the backup process controller takes over for a failed machine, the operation output value directly transferred from the regular process controller by the operation of the switching device is provided. A process controller system is characterized in that a backup operation is started after setting the initial value of the operation output of the process controller system.