JPS5858601A - Split type process control system - Google Patents

Abstract

PURPOSE:To obtain a system having high reliability, by providing each specified inter-subsystem bus, subsystem, in-subsystem bus and selecting signal bus. CONSTITUTION:Inter-subsystem buses 800A-800B formed redundantly by transmission lines of >=2 systems, process input/output subsystems 50, 60 couled with these inter-subsystem buses through an interface, and plural subsystems containing operation control subsystems 20A, 20B are provided. Moreover, each in- subsystem bus for transferring each internal signal of these plural subsystems, and a selecting signal bus for transferring in common selecting signals Slt-A, Slt-B for controlling said interface between these respective in-subsystem buses and said inter-subsystem buses 800A-800B, to each subsystem are provided, by which a split type process control system is constituted.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a so-called Sgerritt-type process control system that centrally controls a process field by function, and particularly aims to provide a highly reliable system.

First, we will explain various conventional systems for four-channel control systems, clarifying that split-type control systems are preferable, and that in the Sgerit-type control system, the means of information exchange within the system is important from the point of view of reliability. Kooa
s+! This invention for Et-enhancement is described.

As shown in FIG. 1, the basic configuration of the process control device is as shown in FIG.
The flow rate signal is a mA analog signal as the process input/output.
The information is input to the operation unit 30 through the Ill500. A control loop t-configuration is configured such that a control signal, for example, an analog valve operation signal of 4 to 20 mA, is transmitted from the operation unit 30 to the process field 5 through a process input/output line 500 t'' to control the process data, that is, the flow rate. Information transmission 400 is performed between the operation unit 30 and the operator 4,
The information includes operation mode switching 410. Operation mode display 420. Setting value change 430, setting value display 440, l
IT fixed value display 450. Output value history 460. Output value display 4
70 etc. The operator performs settings, operations, etc. on the operating section 30 while viewing this information.

The operation unit 30 communicates information 300 with the arithmetic control unit 20, and the information includes operation EI7 monitor bond 31O1 control target f[(2) remote/local mode display signal 320. Setting value change signal 330. Set value answer back signal 34G, operation mode status signal 350 indicating automatic/medium movement. Output value change signal 360 for process output. There are measurement signals 370 representing process quantities. The calculation control unit 20 calculates these signals from the operation unit 30 and outputs an output change note 360 for controlling the process so that the measurement signal 370 becomes the set value 840K. The arithmetic control unit 20 exchanges information with the host system IO through the communication line 210t.

When the arithmetic control unit 20 performs multi-loop control that controls a plurality of single loops shown in FIG. 1, as shown in FIG.
A plurality of operation units 30 are connected to a common calculation control unit 20 through information transmission means 300 using individual signal lines. In this case, when the number of control loops increases, the information transmission means 3
The amount of wiring for 00 becomes enormous, which is a practical problem.

The part indicated by the chain line in FIG. 2 is the process field 5.
1 represents the interface to the digital field 6 corresponding to the analog 1. In the process field, there are also many devices that interface with digital signals such as solenoid valves, solenoid switches, signal lights, operation switches, alarms, and 11/10 bill verification devices. Therefore, it is desirable that the process control device be compatible with both the analog field 5 and the digital field 6. Further, the analog field 5 and the digital field 6, which are differentiated by signal type, have a correlation with each other and are often included in one target process.

In FIG. 2, the digital field 6 is connected to the digital operation i#70 through the digital process input/output signal line 600, and the digital operation device 70 and the arithmetic control unit 20
The information transmitting means 800 and 700 are connected by the information transmitting means 700 using the individual signal line system, and as shown in FIG. Each operating section 30. In some cases, a microcombi unit is incorporated into the digital operating device 70 as a communication means, thereby omitting the large amount of wiring.

FIG. 4 shows a common arithmetic control section 20t-not used, and each operation section 3
This is an example in which the arithmetic and control functions are shared with a microcomputer. In the systems shown in FIGS. 2 and 3, a failure in the common arithmetic control unit 20 would cause the control of all loops to stop, but the system shown in FIG. 4 is superior in terms of risk distribution. There is.

This idea of dispersion is an effective way to maintain continuity of plant operations. For example, analog system control loop 1
When the system breaks down, a portable manual operating device is connected to the process input/output line to serve as a backup manual, and by replacing the failed control a with a spare part, #i operations can be maintained. In the case of simultaneous failures in multiple systems, it is practically impossible to carry out such backup through a large number of personnel, and in this respect the system shown in Figure 4 is better than the system shown in Figure 3. If simultaneous failures occur, it becomes an excellent tool for maintaining operations.

On the other hand, in contrast to this idea of dispersion, it is also possible to maintain operation using a rational method of backup. When pursuing an excellent backup method, it is more advantageous to have a background of concentration rather than decentralization.

In other words, in Fig. 3, the idea is that continuity of operation can be maintained by preparing two calculation 1vl1 parts 20 and switching between them, and this idea is valid when dispersion is intended as shown in Fig. 4. Can not. A system that rationally pursues concentration over decentralization is a system called the so-called split type. FIG. 5 shows this method, in which the intra-system information exchange means 800 is used as a subsystem. l1
lJ m part 20. Operator console 40. Analog human output section 50. A digital input/output unit 60 is interfaced.

If we have such a split type system K>, for example, Fi
As shown in Figure lJE6, the analog input/output section 50 has multiple loop input/output units) 51, 52, 53... songs... 5n is internally input, and the arithmetic control section is put on standby by two 20A and 208. Since it is necessary to construct a redundant system, the individual units 51, 52, 58, . . .
The units 51 to Sm can easily achieve high reliability, and the arithmetic control section 20A, ROB
By allowing the two to perform a mutual backup function,
Plant operation maintenance can be ensured as much as or better than a distributed system.

So-called sequence control systems that control digital fields often perform field control based on a complex sequential system, and the method of directly accessing fields using manual operation is not necessarily effective in maintaining operations.

It is possible to some extent to apply the idea of distributed control to a sequence control system with the intention of dividing it into four systems, but this results in restrictions being placed on the sequence control system and the freedom to form a general-purpose system. The degree becomes smaller. Therefore, automatic backup methods are more suitable for sequence control systems than for analog control systems.

Maintenance methods in distributed systems are based on the premise of prompt response to individual failure points, but in centralized systems, the premise is that operations are maintained by first switching between failed machines and standby machines using an automatic backup system. Next, we request that you take action on the malfunctioning machine. A rounding operator who performs such backup has a large margin of work.

The lifeblood of an excellent split-type system based on centralization and automatic backup is the information exchange means 800. On the other hand, in a distributed system, for example, the information transmission line shown in Figure 4!
10 does not necessarily have an impact on operations. In the split type, it is necessary to arrange such important lines. When the arithmetic control unit s20 and the plurality of operation units 30 in FIG. 2 are connected by individual signal lines, the number of signal lines is too large even if the arithmetic control unit 20t is duplicated and a standby redundant system is used. Switching operations associated with automatic backup become vIiIll. Therefore #! It is desirable that the information exchanger R800 in the split type shown in FIG. 6 has a simple structure.

Hitoshi's invention is based on a split-type process control system, which uses an information exchange means to interface with multiple subsystems such as process input/output subsystems and arithmetic control subsystems as a path between subsystems. In addition, the 1-stage five-way bus is made redundant with at least two transmission lines to connect the subsystems, simplifying the installation and increasing reliability. Furthermore, the interface between the internal bus and the inter-subsystem bus is controlled by the SEL/RI) No. 11 through a select bus commonly connected to each subsystem. By doing K like this, you can easily do this7.
'Intersystem buses can be switched to connect subsystems.

For example, as shown in FIG. 7, the information exchange means 8oo can be arranged redundantly on the inter-system paths 800 and 800B, and the process input/output of the analog input/output section 50, which is the subsystem's 51. S!・・・・・・
・0 input/output bus 500 in the subsystem connected to 5n
0, and the internal subsystem bus 800A, 800 through which it is connected.
B are respectively combined. Although not shown in the figure, there are other subsystems, i.e., the functions shown in Figure 5! Control unit 20. Similarly, the operator console 4G, digital input/output unit 60, etc. are connected to each subsystem via the interface of each bus company within each subsystem.
800B.

In this way, if the system platform Takao prefecture exchange means 8oo is formed with a bus structure to create a simple line configuration and multiple buses are arranged in Marukami to create a redundant structure, the reliability of each bus will be guaranteed. There is no need to attach importance to the transmission method, and it becomes possible to use any type of transmission method.

In the configuration shown in FIG. 7, the most important part that requires reliability is the subsystem internal bus soo.

K is promised. In other words, it is necessary to be able to transmit process information to the normal inter-subsystem bus without being affected by the faulty inter-subsystem path, and when the intra-subsystem path 5000 has a structure that can maintain this, The first inter-subsystem bus 8o.

The multiplexed redundant structure makes it possible to make the L-1 or inter-subsystem bus 800 into any convenient form.

From this point of view, it is preferable that the bus ink 7 ace has a redundant structure for the Kerbus driver channels, and that the channel selection method has a redundant structure.

Generally, the signal lines are arranged in a bus structure, and each pin (Bl, 82, 83) of the signal source is connected to the bus structure as shown in FIG.
Driver switch S□e B@@ * 86@K can be connected in series to KII and connected to bus 90 in series, or alternatively, as shown in FIG. 8B and C, There is a parallel switch system in which bus driver switches S□, S□, and S□ are connected in parallel.

To make the bus driver in the interface have a redundant structure, for example, as shown in FIG. ends to the common potential point KgI! through the parallel switches stlstm #SPm, respectively. Continue. Both the serial format and the parallel format of trench @8 diagrams are used. Parallel type switches S□, S□.

8p1 may be of a type driven with mutually opposite polarities as shown in FIG. 9. By using both series switches and parallel switches in this way, a set of parallel switches with reverse polarity drive can be created as shown in Figure 10.
Channel selection is possible by switching between the power supply and the common potential point using the common channel selection switch Sc of each power supply at-2 of $s,, #Bl. Further, in this example, a diode switch is used as the series type switch 811 #S■e881 to automatically generate a switching action.

In addition, as shown in Figure 11, two channel selection switches S (11801) are connected in parallel for redundancy, and each series switch 814 * sss t s is connected to the diode series switch 5lly 811 t 811 t'.
New effects can be generated by using si in combination.

Parallel redundancy of the wholesale channel selection switch SCI t sag increases the failure probability in the short axis direction and lowers the failure probability in the open direction, and series redundancy of series type switches increases the failure probability in the open direction. This increases the probability of failure in the wing direction. This means that the probability that the failure effects of these switches will go outside, that is, to the bus 90, is small;
The problem is that the failure effect is more likely to be directed internally, that is, when this bus driver fails, the bus 90
If the polarity of the failure is selected in a mode that does not exclusively use the

A bus driver that uses the redundancy system that selects the fault polarity in this way is, for example, ll5! As shown in the figure, when two channels Chr and Chs share the bus 90, there is an effect that a failure of one channel does not interfere with the bus driver of the other channel. 'In addition, in Figure 4111 and Figure 12, the channel selection switch Sc is the selection signal Bit I.
The IC activates that channel.

FIG. 13 shows an example of the combination L1 of the inter-subsystem bus shown in FIG. 7 and the intra-subsystem bus LIR using the interface having the bus driver shown in FIGS. 10 to 12. Interfaces 520A and 520BK are connected to inter-subsystem buses 8GOA and 800B, respectively, and control signals 1ml for supplying selection signals 8jtA and 8AtB, respectively, for activating the interfaces.
is connected. Also, the interface 520A,
511G11 is also connected to select bus 908jt,
Selection signals Btt 51, 81162, . . . 8, which selectively activate the process input/output crab units 51, 52, . . ., 5m through this selection bus 9o stt.
jt5nt? Each is supplied separately.

The activation of not only the interfaces 520 and 520B but also each process input/output unit is controlled by the fault polarity, and is then connected to the bus by the redundancy method. Therefore, a failure of one of the inter-subsystem buses 800A and 800BD is a threat. The other input/output crab units 51...5n can be connected to the subsystem internal bus 5ooo without any connection, and regardless of one or more of the input/output crab units 51...5n, other input/output crab units 5n can be connected can normally input and output information to and from bus 5000.

reading data on the intra-subsystem bus 5000;
Reading back can be done arbitrarily, for example by connecting a buffer Br to the bus 5000 through a series resistor Rb to form a receiver channel as shown in Figure IE14. The purpose of the third series resistor Rh is to prevent a failure of the buffer Br from spreading to the bus 5000.

In Figure 13, there are 520 or 52 interfaces.
0B reads back outgoing data from the intra-subsystem path 5000 and sends the data back to the inter-subsystem bus 800A or 800B. If it is possible to monitor the failure of the driver channel to the internal bus 5000, the selection signal 84tA is set based on the monitoring result.
, it is possible to operate 8AtB4.

Bus maintenance using the series resistor Rb used in the receiver channel in Figure 14 is performed by combining it with the driver channel as shown in Figure 1S, applying an analog input to the non-inverting input terminal of the operational amplifier Pass the input through resistor Rbt to bus 5o
oOK'I can be used to form an analog driver channel by connecting the output end to a series switch SSt and forming a path of 5000km1.

Maintaining the functionality of the data bus between each subsystem that makes up a split system is important for maintaining the functionality of the entire system, and redundancy of the subsystem related bus 800 is also based on the integrity of the intra-subsystem path 5ooo. I have stated above that this is an established thing. In FIG. 13, the $P loose selection signals 5ztA, 8ttB correspond to 20 people, 20B, 40°80,
60 may be a common signal. This selection signal 8j
The tA, 5ttB Fi selection signal generation section 80 generates the signal, and the generation may be performed manually using a simple changeover switch, or may be operated using an output signal from a host system.

The integrity of the bus (5000) within each subsystem is determined by the 4I image in its bus structure and selection notes 8ztA s 5
The selection signals 5AtA and 8jtB are maintained by using external signals, and the signal form of the selection signals 5AtA and 8jtB should be as simple as possible, for example, a simple DC level signal. It is possible to superimpose this signal and to maintain maximum reliability.

The parts to be switched by the selection signals 5jtA and 8ttB are inter-subsystem buses 8GOA and 800B, and bus interface parts within each subsystem, for example 52.
O.A.

520B, the switching of this interface part gives priority to the selection of each interface part itself, and it is connected only when the selection signal is received, and the tC that remains connected due to a failure is connected.
It will be harmonized so that it does not happen.

As described above, according to the present invention, a split type system is used, and information transmission between the subsystems is configured as a bus, and by making the bus redundant, the selectivity of the information transmission format is large, and the m1fit of the transmission path is The r4 reliability can be significantly increased only if it is simple.

By providing component redundancy to the interface driver so that the probability of a failure that continues to output power to the other bus is smaller than the probability of a failure that stops outputting power to the other bus, it is possible to prevent failures from affecting the bus. This makes it possible to further improve reliability. Furthermore, reliability can be improved if the analog or gage digital signal is switched using a diode switch. For example, with a diode, 5F'iT (1/(5X10
However, transistors have a 100 FIT failure rate, and diodes have fewer failures, and reliability can be improved by using diodes.

[Brief explanation of drawings]

Figure 1 shows the pull 7th control system. Fig. 2 is a block diagram showing a nine-process control system in which multiple loops are combined with a common arithmetic control section, and Fig. 3 is a block diagram showing the basic configuration of each operation section in Fig. 2 and a common arithmetic control system. Fig. 4 is a block diagram showing a process control system in which parts are connected in a bus structure; Fig. 4 is a block diagram showing a distributed process control system;
The figure is a block diagram showing a split-type process control system. Figure 6 is a block diagram showing nine examples of duplicating subsystems in a split-type process control system.
Fig. 47 is a block diagram showing an example of a split type process control system according to the present invention, Fig. 8 is a connection diagram showing a general example of each S of the interface bus driver, and Fig. 9 is a diagram showing a general example of the bus driver of the interface. Connection diagrams showing examples of bus drivers for Ink 7Ace, Figures 10 to 12 Connection diagrams for various types of interface bus drivers applicable to the system of the present invention, and 0St-Boss connection diagrams, Figure 13
Figure 14 is a diagram showing the structure of the unit selection signal part in this system, Figure 14 is a diagram showing fl of the receiver part of the interface, Figure 15 is a connection diagram showing the analog signal bus driver, FIG. 16 is a block diagram illustrating a general implementation of a split process control system according to the present invention. 20A, 20B: Behind-the-scenes control subsystem, 30:
Operation unit, 40: Operator console, 50: Analog clock input/output subsystem, 60: Digital process input/output subsystem, 80A, 110B: Inter-subsystem bus, 5ooo: Intra-subsystem bus, 5L
-5n: In-out crab knit, 520A, 520B:
Interface, 5AtA, 8jtB: Selection signal, 90 BLt: Selection signal, S□v sss
# sag: Series switch, srs e 8Pj
w srs: parallel type switch, 81+=selection switch. Patent applicant Hokushin Electric Manufacturing Co., Ltd. Agent Kusano
Wisdom 3 Wisdom 4 (Wealth) Sai 5 Wisdom 6 Mouth 7 7 Goodness 73 Wisdom 9 圀÷F Sai 10 Mouth

Claims (5)

[Claims] (1) A redundant inter-subsystem bus with at least 42 transmission lines, at least a subsystem input/output system, and an arithmetic control system connected to these intersubsystem paths via an interface. a plurality of subsystems including At, each intra-subsystem bus for transmitting internal signals of each of these plurality of subsystems, and a selection signal that controls the interface between each of these intra-subsystem buses and the inter-subsystem path; A split-type process control system with a selection signal path that commonly communicates to each subsystem. (2) The interface includes a channel selection switch that outputs power W4 according to an external selection signal, a parallel switch means that uses the output of the channel selection switch as a power source, and an output signal of the O parallel switch means that opens and closes the output signal. 9. The split type process control system according to claim 1, further comprising a serial type switch and switching means for transmitting information to said subsystem internal bus. (3) The channel selection switch discharges power t
The split-type 10-cess control system according to claim 21i1, wherein components are arranged redundantly so that the probability of failure in a single unit is smaller than the probability of failure in a situation where no power is output. 0 (4) The split-type process control system according to claim 21i1, wherein in the interface, the series switching means has components redundantly arranged so that the failure probability of an open circuit is smaller than the failure probability of a short circuit. . (5) The split type process control system according to claim 4JJ, wherein the series type switching means is formed of a diode circuit. (b) The inter-subsystem path is connected to the @1 inter-subsystem bus and the second subsystem bus,], and the previous W-in-p Fenis connects the @10 subsystem bus and the intra-subsystem path. a first interface connecting the second inter-subsystem path and the intra-subsystem path; and a second interface connecting the second inter-subsystem path and the intra-subsystem path;
2. A split-type device as claimed in claim 1, wherein the units are configured such that the bus driver channel is activated by the enit selection code output from the first or g2 interface. Process control system.