JPH0865323A - Data transmission system for distributed control - Google Patents

Abstract

PURPOSE: To efficiently use a transmission line by outputting a synchronizing signal to the data transmission line for a prescribed time from the end of the frame at the time of the end of the transmission/reception processing for one frame in each node. CONSTITUTION: A communication processor 30 transfers data received from another node to a control processor 40 and monitors a transmission line B; and when it is the timing of the transmission line use period of its own node, the processor 30 outputs data received from the control processor 40 to the transmission line B. Meanwhile, the control processor 40 transfers data related to apparatus 11 and 12 to the communication processor 30 based on this received data for the purpose of transmitting it to another mode. In the communication processor 30, a synchronizing signal output means 33 outputs the synchronizing signal to the transmission line B after receiving arm data from another node. A frame start point detection means 34 monitors the time elapsed after disappearance of all the synchronizing signal on the transmission line and detects the start point of the frame. A transmission line use period determining means 35 autonomously determines the transmission line use period.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmission line sharing technique for a distributed control system using a bus type communication network.

[0002]

2. Description of the Related Art In the control of a system composed of a plurality of physically distributed devices, a computer for processing signal input / control output is arranged for each device and these computers communicate with each other. A form of so-called distributed computer control system (hereinafter simply referred to as a distributed control system), which controls the entire system while connecting to each other through a network and exchanging information with each other, becomes a centralized control system in which all devices are controlled by one computer. Compared to,
It has come to be used in a wide range of fields because it is advantageous in terms of reducing the number of wires, fault tolerance, and expandability.

As a connection form (topology) of this communication network, there is a typical type generally called a mesh ring bus or the like, but for system control,
The mesh is not suitable for the wiring cost, and the ring has a transmission delay at the processing node that relays (a unit of the communication entity connected to the network, hereinafter simply referred to as a node), so it is a bus type in terms of ensuring real-time performance. Inferior.

On the other hand, the bus-type network has an advantage that all nodes can receive the signal output from each node immediately and simultaneously, and the bus-type network is suitable for a distributed control system requiring real-time property. .

FIG. 19 is a diagram showing an example of the overall configuration of a distributed control system constructed by this bus type network.

[0006] In this example, four nodes # 1 to # 4 are connected to a bus type data transmission line B, and each node exchanges information with each other, and each of the control target devices 11 to 16 connected to its own node. Shows the configuration for controlling the entire system by controlling the.

[0007]

By the way, when using this bus type network, one of the biggest problems is how to make a plurality of nodes share this "data transmission line = bus". This is so-called transmission line sharing technology.

A polling method has been well known as a transmission path sharing technology for a bus type network. The state of signal processing when this polling method is applied to the configuration of FIG. 19 is shown in the time chart of FIG. explain. Here, it is assumed that the node # 1 is a control station (host station) and the nodes # 2 to # 4 are subordinate stations (slave stations).

[0009] Between the star and the star in Fig. 20, that is, from the start of data output of the node # 1 to the end of data output of the node # 4 is called a transmission frame (hereinafter simply referred to as a frame), and various numbers in parentheses are used. Time required for processing (unit: ms
ec) is shown.

First, the node # 1, which is the control station, first outputs its own data at an arbitrary time. Then # 2
~ Sequentially poll nodes # 4 (call origination, P # 2-
P # 4). Here, the "transmission path use period" required for each node to output data is 50 msec, and the time required for polling is 5 msec.

Then, the nodes # 2 to # 4 judge the timing of the transmission path use start of their own node according to the polling from # 1.

In this way, each of the nodes # 1 to # 4 sequentially outputs data to the transmission path, and when reception of all data from the other nodes is completed (at the frame end),
The above process is repeated again.

The polling method exchanges information between the nodes as described above, but has the following problems.
That is, 1. Additional communication is required for the control station to poll. 2. Due to the failure of the control station, data transmission / reception of the entire system becomes impossible. 3. A method has also been proposed in which a failure of a control station is monitored by a timeout and one of the remaining stations automatically acquires the control right, but even in that case, the real-time property is temporarily lost. 4. The failure of the dependent station will be monitored by time-out, and the real-time property cannot be maintained at least temporarily. However, there are problems in terms of transmission efficiency, fault tolerance, and real-time performance.

In addition to the polling method, a token method, a contention method, etc. are known. Additional communication, such as sending tokens, is required. 2. Due to the failure of the node holding the token, data transmission / reception of the entire system becomes impossible at least temporarily. As such, there are still problems in terms of transmission efficiency, fault tolerance, and real-time performance. In the case of the contention method, 1. Even if the average communication capacity is sufficient as a whole, waiting time occurs when the transmission timings collide. Therefore, the transmission delay cannot be quantitatively defined, and it is not suitable for a control system that requires real-time processing. There is such a problem.

Therefore, an object of the present invention is to enable communication to be continued between the remaining nodes without losing the real-time property even when some nodes fail. An object of the present invention is to provide a data transmission system that is excellent in fault tolerance, real-time performance, and transmission efficiency, such as not requiring an additional transmission line sharing sequence.

[0016]

In order to achieve the above object, the present invention is configured as follows. That is, a plurality of processing nodes are connected to a bus-type data transmission line, and each processing node exchanges information with each other via the data transmission line, thereby forming a system configured by a group of devices connected to each processing node. In the distributed control system configured to control the whole, each of the nodes transmits a synchronization signal for each node to autonomously perform communication synchronization after one cycle of transmission / reception processing to the data transmission path for a predetermined period of time. A synchronization signal output means for outputting to the above, a frame start point detection means for detecting a start point of the next frame by monitoring an elapsed time from the time point when all the synchronization signals of the respective nodes in the data transmission line become no signal, The configuration is provided with a frame end from the frame start point and a transmission path use period determining means for autonomously determining the transmission path use period of the own node.

[0017]

In the data transmission system having such a configuration, when each node completes the transmission / reception processing of one frame,
A synchronization signal is output from the end of the frame to the data transmission line for a predetermined time.

Further, each processing node monitors the elapsed time from when all the synchronizing signals on the transmission line have become null, and detects the starting point of the next frame. Each processing node autonomously determines the transmission path usage period of its own node from this frame start point by comparing it with a value given in advance for each node, and at that timing, it transmits data and other nodes Receives it as data from that node.

[0019]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing an example of the configuration of a processing node according to the present invention, and shows the case where the present invention is applied to the distributed control system having the configuration of FIG. 19. The same parts as those in FIG. It is shown in. In addition, # 2 to # 4
The processing node of No. 1 has the same configuration as that of # 1, and is not shown.

As shown in FIG. 1, each node has
A communication interface 20, a communication processing device 30 for transmitting and receiving data, a control processing device 40 for exchanging information with other nodes via the communication processing device 30 and for controlling the controlled devices 11 and 12. An input / output port 50 is provided.

Further, the communication processing device 30 synchronizes after receiving all the data from the transmission processing means 31 for performing the transmission processing to the transmission path, the reception processing means 32 for performing the reception processing from the transmission path, and the other nodes. A sync signal output means 33 for outputting a signal to the transmission path, a frame start point detection means 34 for detecting the start point of the frame by monitoring the elapsed time after all the sync signals on the transmission path have become no signal, and a frame start point for the frame It is composed of a terminal and a transmission path use period determining means 35 which autonomously determines the transmission path use period of its own node.

Here, the term "no signal" means that the node outputs when the configuration as shown in FIG. 2 is adopted (open collector inverter is used for output and inverter is used for input).
It means the state of disable (= high impedance).

When all nodes have no signal (due to pull-up resistors), "0" is obtained as received data at each node.

When the node outputting "1" and the node outputting "0" overlap, "1" is set in all nodes.
The circuit is configured to be recognized as.

In the above configuration, the communication processing device 30 delivers the data received from another node to the control processing device 40, and also monitors the transmission path and when the timing of the transmission path use period of the own node comes, the control processing device 40. The data received from is output to the transmission line. On the other hand, the control processing device 40 gives a control command to the devices 11 and 12 based on the data, or performs a process of handing over the data to the communication processing device 30 in order to transmit the data related to the devices 11 and 12 to another node. .

FIG. 3A shows the synchronization signal, and FIG.
Data (data block) output by each node
An example of each format is shown below.

The synchronization signal is a signal which is desired to output data and which can be output, and which is output from all the nodes to the transmission line all at once after all the nodes have transmitted the output data.
As shown in the figure, the output is terminated after the data block length has passed at least. When all the nodes are in the output disabled state (disabled), the transmission path is in the no signal state, that is, the state of "0".

Each processing node monitors the elapsed time from this no-signal state, determines the starting point of the next frame, and autonomously determines the transmission path use period of its own node from the starting point of the frame by the method described later. . Then, at the timing of the start of the transmission path use period of the own node, the data block shown in FIG. 3B is output.

As shown in the figure, the data block is composed of a head (header) of the bit pattern "1", data of an arbitrary bit string, and an end (trailer) of the bit pattern "0".

Therefore, it can be clearly distinguished from the synchronization signal in which "1" is continued for the data block length or more.

Further, as shown in FIG. 3 (e), a sync signal having a continuous "1" bit pattern having a length of n bits or more is used, and the transparency of this bit pattern is maintained in the data block. Even if this bit pattern does not appear, the sync signal can be detected. An algorithm for preventing this pattern from appearing in the data block is described in, for example, HDLC (JIS-X-5104) and is not directly related to the present invention, so its details are omitted.

Next, in order to realize the above operation, the procedure of processing of the communication processing device in each node will be described with reference to the flowcharts of FIGS. 4 to 6 (common to each node). (In the following description, FIG. 3 (a) is used for the synchronization signal and FIG. 3 (b) is used for the data block.) In FIG. 4, the node enters the transmission path after the power is turned on. It shows the procedure up to.

First, when the power is turned on, the communication processing device sets the output disable state so as not to affect the transmission path until the communication control is started (step S11). Then, in order to monitor the transmission line in that state, a decrement timer for entry monitoring is preset to an initial value (a period corresponding to one transmission / reception cycle, that is, a period corresponding to the start point of the next frame). Step S1
2). Then, in steps S13 and S14, it is monitored whether or not the transmission path is in the "1" state within this period.

If the transmission line has not been used at all yet, the transmission line does not become "1" within this period, so if the time-out occurs in step S14, the process proceeds to the main loop.

On the other hand, when the transmission line is already used by another node, there is a moment when it becomes "1" by the data block or the synchronizing signal of those nodes within this period. A decrement timer for monitoring the synchronization signal is used to check in steps S15 to S17 to detect that the signal is the synchronization signal. Next, in step S18, it is determined that the value is “0” during the starting point monitoring period −Δt.
~ Check with S1A. Here, Δt is a margin for preventing a judgment error due to a subtle difference in frame origin recognition of each node, and is set to, for example, 0.1 msec. Furthermore, actually 1
It is checked in step S1B whether the data for the frame can be normally received, and if normal, the process proceeds to the main loop.

If it is abnormal, the process returns to step S12 to try to enter again.

FIG. 5 is a flow chart showing the processing procedure in this main loop. As described above, the transmission line is entered for the first time after the power is turned on, or is already in the transmission line and the transmission / reception process described later (step S2A). When the process ends and the end of the frame is reached, the sync signal is output for the preset time of the sync signal output decrement timer in steps S21 to S25.

If any one of the other nodes outputs the synchronization signal, the transmission line remains "1" during that period, but when the synchronization signal output of all the nodes is completed, Also becomes "0". When this is detected in step S26, each node simultaneously monitors the elapsed time from this point (steps S27 to S29), and when the decrement timer immediately before the frame start point determination times out, each node simultaneously recognizes the point immediately before the frame start point. Then, each node autonomously determines the transmission path use period of its own node from the start point of the frame by the transmission / reception process of step S2A as described later, and outputs data at that timing.

On the other hand, while the frame origin point is being monitored and the transmission line must be "0", it is actually "1" (step S28), or an abnormality has occurred during transmission / reception processing (step S2B). In case of abnormality such as step S
The process proceeds to 2C, and the jam sync signal having a longer output time than the normal sync signal is output even in the middle of the frame. As a result, communication abnormalities also occur in all the other nodes, and the synchronization signal for jam is output in the same manner as the node that first detected the abnormality.

In this way, since all the nodes output the synchronization signal (for jamming), the frame starting point can be recognized at the same time by the above-described procedure, and the normal communication state can be quickly returned.

FIG. 6 is a flow chart showing details of this transmission / reception processing.

First, in step S31, the current node number is set to 1. Then, through step S32, S33
Then, it is determined whether or not the current node number is equal to the own node number. At this time, in the processing node of # 1, step S3
From S3 to S34 and S35, the data of the own node (the data block of FIG. 3B) is transmitted with the predetermined occupation time (for example, data output 50 msec + no signal 1 msec 51 msec in total) as the transmission path use period. Here, Δt is a margin for preventing a misjudgment due to a subtle difference in the frame starting point of each node (for example, 0.1 msec) and corresponds to the time from immediately before the frame starting point to the frame starting point. And step S3
In step 7, the current node number is incremented, and the process returns to step S32.

On the other hand, the nodes # 2 to # 4 have different current node numbers and own node numbers.
The process proceeds from step S36 to step S36, and the data is received as data from the node # 1 during the transmission path use period. At step S37, the current node number is incremented and then step S32 is performed again.
Return to.

Then, the current node number is 2 this time.
Therefore, in the node # 2, steps S33 to S3 are performed.
4, the process proceeds to S35 to output the data, and at other nodes, the process proceeds from step S33 to S36 and the data is transmitted to # 2.
Received as data from the node.

After that, the same procedure is repeated, and when the current node number becomes 5 and becomes larger than the maximum node number, transmission / reception of one frame is completed, and step S in FIG.
Return to 21.

After that, the procedure of FIG. 5 is repeatedly executed, whereby information transmission / reception processing is performed between the nodes via the transmission path.

FIG. 7 is a flow chart showing the details of the frame check process in step S1B of FIG. 4 (the same symbols as those in FIG. 6 represent the same processes). This is a transmission / reception process (Fig. 6)
Since the transmission processing is omitted from the above, the description thereof will be omitted.

FIG. 8 is a timing chart showing the processing contents of each node in the above procedure, the output to the transmission line, and the signal state of the transmission line.

Each node outputs a sync signal from the end of the frame, and then 5 from the time when all the sync signals become null, that is, the time when the transmission path becomes "0".
The frame start point is determined after msec has elapsed.

From the frame origin, the node of #N is (N-
1) Start using the transmission path after 51 msec, 51x4 =
The figure shows a state in which the transmission / reception process is terminated when 204 msec has elapsed, and the synchronization signal is output again in preparation for the next transmission.

The operation when all the nodes # 1 to # 4 are normal has been described above. In FIG. 9, when the nodes # 2 and # 3 fail and communication processing becomes impossible, Alternatively, the total number of nodes designed as a system is 4, and the ones installed at a certain site are # 1 and # 4.
A timing chart is shown when there is only one node.

As is clear from FIG. 9, even when the communication processing cannot be performed due to a failure of the nodes # 2 and # 3, the nodes # 1 and # 4 perform the communication processing in the same manner as in the normal case. It can be carried out. That is, according to the above embodiment, # 1
Node determines the next frame starting point from the elapsed time after the transmission path becomes “0” after the frame termination, and starts using the transmission path 51 × (1-1) = 0 msec after the frame starting point. Similarly, the # 4 node also detects the frame origin,
51 × (4-1) = 153 msec later, each node autonomously determines the period of use of the transmission path from the frame start point, such as starting the use of the transmission path and transmitting data. Communication processing can be performed regardless of the failure or existence of the node.

By the way, according to the above embodiment, since there is no polling sequence or token transfer sequence by the control station, not only can the transmission path be used efficiently, but
Even if the function of an arbitrary node stops due to a failure, etc., communication between other nodes will not be temporarily stopped, and transmission delay will not occur, and communication will continue exactly as in normal times. However, as is clear from FIG. 9, if there is a node that does not send data due to a failure or some other reason, the transmission line use period corresponding to that is wasted.

A second embodiment in which this point is improved will be described below.

FIG. 10 is a timing chart in the second embodiment and corresponds to FIG. Here, the transmission path use period of each node is (N−1) × 51 from the frame start point.
Each node autonomously determines not by msec but by the accumulated time of the no signal state after the frame start point. That is, assuming that the unit time for detecting no signal is 1 msec, the node of #N has (N-1) × the no signal accumulated time from the starting point of the frame.
Start using the transmission path when 1 msec has elapsed.

The end of the frame is set at the time when the accumulated time of no signal from the frame starting point becomes the total number of nodes × 1 msec, that is, 4 msec in this case, and the operation from the frame ending to the detection of the next frame starting point is the first. It is exactly the same as the embodiment.

FIG. 11 is a time chart when the nodes # 2 and # 3 do not output data for some reason in the second embodiment, and is equivalent to FIG.

As is apparent from FIG. 11, since each node determines the transmission channel use period based on the accumulated time of no signal from the frame start point, for example, the node # 4 outputs data after the node # 1 outputs data. Data will be output after 3 msec has passed, that is, the idle time of the transmission path in one frame is always the number of all nodes x 1 msec regardless of the number of failures of the node. Waste can be greatly improved.

FIG. 12 shows the procedure of transmission / reception processing for realizing the second embodiment, which corresponds to FIG. 6 of the first embodiment.

First, when the point immediately before the frame start point is determined, the no-signal measurement timer is preset to the initial value (-Δt) in step S81, and the transmission / reception flag for indicating whether transmission / reception has been performed is reset in step S82.

Next, f1 is set to the current transmission node number (node number to be currently transmitted), and f2 is set to the current reception node number (currently received node number) (steps S83 and S84).

Here, f1 and f2 are calculated as follows.

F1 = Int [(value of no-signal measurement timer /
No signal measurement unit time) +1] f2 = Int [(No signal measurement timer value + Δt) / No signal measurement unit time + 1] However, Int [x] = maximum integer not exceeding x Δt = each node The margin for preventing a judgment error due to a subtle difference in the recognition of the frame starting point is, for example, 0.1 msec.

Therefore, for example, f1 = 1 when the accumulated time of no signal after the frame starting point is 0 or more and less than 1 msec, and f1 = 2 when 1 ms or more and less than 2 msec. On the other hand, f2 is earlier than that by Δt, and f2 = 1 when the accumulated time of no signal is more than -0.1 msec and less than 0.9 msec.
When 0.9 msec or more and less than 1.9 msec, f2 =
2 and so on.

Therefore, at the first time point (immediately before the frame start point), f1 = 0 and f2 = 1 and the current transmission node number is smaller than the maximum node number (4 in this case), so the flow proceeds from step S85 to S86. The current sending node number is compared with its own node number.

Since the current transmission node number is 0, any node proceeds from step S86 to step S92.
Here, due to a subtle difference in the frame origin recognition of each node, even if data arrives before the time when the node recognizes the frame origin, the current receiving node number is 1, so that the receiving process after step S93 described later is performed. Can be received as data of the node # 1.

At the time point after Δt (frame start point), f1,
Since both f2 are 1, the process proceeds from step S85 to S86 in the same manner as above, and the current transmission node number is compared with its own node number.

Since the current transmission node number is 1, the node # 1 has steps S86 to S87.
However, since the transmission / reception flag is still 0, that is, no transmission / reception, the no-signal measurement timer is temporarily stopped in step S88, and the data of the own node is transmitted (step S89).
After that, the no-signal measurement timer is restarted (step S90),
The transmission / reception flag is set to 1 (step S91).

After that, while the current transmission node number is 1 (accumulation time of no signal is less than 1 msec), the process proceeds to steps S85 to S87 in the same manner as above, but the transmission / reception flag is 1, that is, transmission / reception has been completed. No processing, S9
After 2 returns to S83.

On the other hand, in other nodes, the process proceeds from step S86 to S92, and when the transmission of the data from the node 1 is started, the transmission path is in the state of "1".
Although the process proceeds from S92 to S93, since the transmission / reception flag is still 0, that is, no transmission / reception, the signalless measurement timer is temporarily stopped in step S94, and the current reception node, ie, # 1
After receiving as data of the node (step S95), the no-signal measurement timer is restarted (step S96), and the transmission / reception flag is set to 1 (step S97).

At this time, even if there is a slight deviation between the timing of the temporary stop of the timer in step S88 and the timing of the temporary stop of the timer in step S94 in another node, the transmission time and the reception time are equal. Therefore, the sum of the transmission / reception time and the no-signal time, that is, the recognition of the occupied time of each node is the same in all the nodes, and the synchronization does not shift.

On the other hand, there is a time lag in recognizing the frame start point of each node, even if it is very small. May occur. The margin for preventing this is Δt described above.

Thus, the above procedure is repeated, and when all nodes complete transmission / reception and the no-signal accumulated time exceeds 4 msec, or even when the communication is impossible due to a failure of any node, the no-signal accumulated time is increased. When it exceeds 4 msec, the current transmission node number becomes larger than the maximum of 4 and one frame of communication is terminated, and as described with reference to FIG. 9, the transmission line can be used efficiently even if a failure of any node occurs. You can

FIG. 13 is a flow chart showing details of step S1B in FIG. 4 and the frame check processing (the same reference numerals as those in FIG. 12 represent the same processing). This is the same as the transmission / reception process (FIG. 12) except that the transmission process is omitted.

In the above, an example in which the period of use of the transmission line is determined for each node and data is transmitted for each node has been described. An example of the case of determining the period of use of the transmission line for each data Will be described.

FIG. 14 is a time chart of the third embodiment, which corresponds to FIG. 10 of the second embodiment. In this example, the total number of data designed as the system is D
It is assumed that there are 100 from 1 to D100, and only the following eight types are used in this system, and the correspondence relationship with the node that outputs each data is as follows.

Node # 1 ... D41 Node # 2 ... D22, D82 Node # 3 ... D13, D53, D63 Node # 4 ... D34, D74 Each node outputs one type of data. It is assumed that the time required for outputting is 10 msec. Also in this embodiment, the operation from the end of a frame to the detection of the starting point of the next frame is exactly the same as that of the above-mentioned embodiment.

As shown in FIG. 14, the transmission path use period of each data is determined by the accumulated time of no signal after the frame start point. That is, the start of using the transmission path of the Dn data is
The accumulated time of no signal after the frame start point is 1 msec x (n-
1) The time has elapsed.

For example, the data of D34 is output from the node # 4 when the accumulated time of no signal from the frame starting point becomes 1 × (34-1) = 33 msec. At the end of the frame, the accumulated time of no signal after the frame start point is 1 msec.
× It is the time when the total number of data, that is, 100 msec.

The procedure for realizing this is shown in the flowchart of FIG.

As in the case of FIG. 12, first, when the point immediately before the frame starting point is determined, the signalless measurement timer is preset to the initial value (-Δt) in step S101, and step S1
At 02, the transmission / reception flags corresponding to all the data designed as the system are reset.

Next, f1 is set to the current transmission data number (type number of data to be currently transmitted), and f2 is set to current reception data number (type number of data currently received) (steps S103 and S104). Here, f1 and f2 are the same as those described in FIG.

As described above, at the first time point (immediately before the frame starting point), f1 = 0 and f2 = 1 and the current transmission data number is smaller than the maximum data number (100 in this case). Therefore, from step S105 to step S106. Then, the current transmission data number is compared with the data number transmitted by the own node.

Since the current transmission node number is 0, any node proceeds from step S106 to S112. Here, due to a subtle difference in the frame origin recognition of each node, even if the data arrives before the time when the node recognizes the frame origin, the current receiving node number is 1, so that the reception processing after step S113 described later is performed. By D
It can be received as 1 data.

After Δt, that is, both f1 and f2 are 1 at the time of the frame starting point, and similarly to the above, step S
The process proceeds from 105 to S106, and the current transmission data number is compared with the data number transmitted by the own node.

The current transmission data number is 1, and
Since no data has been output to the transmission line yet, the process returns to step S103 through steps S106 and S112. By repeating this, when the current transmission data number becomes 13, the node # 3 is the data number transmitted by its own node, and therefore the process proceeds from step S106 to S107, and the transmission / reception flag of that data number is still 0, that is, not transmitted. Therefore, in step S108, the no-signal measurement timer is temporarily stopped, the data of D13 is transmitted (step S109), the no-signal measurement timer is restarted again (step S110), and the transmission / reception flag of the current transmission data number is set to 1. (Step S111).

On the other hand, in the other processing nodes, the process proceeds from step S106 to S112, and when the transmission of the data from the node # 3 is started, the transmission path is in the state of "1", so that the steps S112 to S113. However, since the transmission / reception flag is still 0, that is, no transmission / reception, the non-signal measurement timer is temporarily stopped in step S114, and the current reception data number, that is, the data of D13 is received (step S115), and the no-signal measurement timer is set. It restarts (step S116) and sets the transmission / reception flag of the current reception data number to 1 (step S117).

In this way, the above procedure is repeated, and when the current transmission data number becomes larger than 100, which is the maximum, the communication of one frame is terminated.

According to this embodiment, since each node autonomously determines the transmission path use period of each data by the accumulated time of no signal from the frame origin, only a part of data is output. However, as in the second embodiment, not only is the transmission path used efficiently, but also by determining the transmission path usage period for each data, it is possible to obtain the freedom of what kind of data is output to which node.

FIG. 16 is a flow showing the details of the frame check process in step S1B of FIG. 4 (the same symbols as those in FIG. 15 represent the same processes). This is the same as the transmission / reception process (FIG. 15) except that the transmission process is omitted.

When the use period of the transmission line is determined for each data, the transmission line can be used efficiently if it is decided by using the accumulated time of no signal as described above. 6 and FIG. 8, the transmission path use period of each data is set to 11 msec (data output 10 msec + no signal 1 msec), and the nth data Dn is (n-1) ) × 11 msec, the transmission is started, and the elapsed time from the frame start point can be determined by comparing it with a value given in advance for each data type.

According to the second and third embodiments, the transmission line use efficiency is improved when there is a node that does not send data due to a failure or other reasons, but the transmission line use period of each node or each data type is kept constant. Therefore, it is necessary to match the one that uses the transmission path for the longest, and in the case of a node or data type with a small amount of data, use of the transmission path is wasted.

Further, since the synchronizing signal requires a time longer than the transmission path use period of each node, this time is also wasted.

A fourth embodiment in which this point is improved will be described below.

FIG. 3C shows the synchronization signal, and FIG.
Data (data block) output by each node
An example of each format is shown below.

The synchronization signal is a signal which is desired to output data and can be output, and which is output from all the nodes to the transmission path all at once after all the nodes have transmitted the output data.
As shown, at least 2 bits or more consecutive "1"
Each node ends output after a stipulated time has elapsed. When all the nodes are in the output disabled state (disabled), the transmission path is in the no signal state, that is, the state of "0".

Each processing node monitors the elapsed time from this no-signal state, determines the starting point of the next frame, and autonomously determines the transmission path use period of its own node from the starting point of the frame by the aforementioned method. . Then, at the timing of the start of the transmission path use period of the own node, the data block shown in FIG. 3D is output.

As shown in the figure, the data block is composed of the beginning (header) of the bit pattern "10", data consisting of an arbitrary bit string, and the end (trailer) consisting of the bit pattern of "01111110". The data part maintains the transparency of "01111110" so that this bit pattern does not appear in the data, and the end can be detected reliably. An algorithm for preventing this pattern from appearing in the data part is, for example, H
Since it is described in DLC (JIS-X-5104) and is not directly related to the present invention, its details are omitted.

Next, in order to realize the above operation, the processing procedure of the communication processing device in each node will be described with reference to the flowcharts of FIGS. 17 and 5 (common to each node).

FIG. 17 shows a procedure from when the power of the node is turned on to when the node joins the transmission line.

First, when the power is turned on, the communication processing device sets the output disable state so as not to affect the transmission path until the communication control is started (step S51). Then, in order to monitor the transmission path in that state, a decrement timer for entry monitoring is preset to an initial value (a period corresponding to one cycle of transmission / reception, that is, a period corresponding to the starting point of the next frame). Step S
52). Then, in steps S54 and S55, it is monitored whether or not the transmission path is in the "1" state within this period.

If the transmission line has not been used at all yet, the transmission line will not become "1" within this period, and if a timeout occurs in step S55, the process proceeds to the main loop.

On the other hand, when the transmission line is already used by another node, there is a moment when this node always becomes "1" due to the data block or synchronization signal of those nodes. Check it in step S53,
A trailer having a unique bit pattern is detected from the data flowing through the transmission line.

The bit pattern that appears next to the trailer is
Since it is either the header or the sync signal, this is checked in steps S57 to S59, and if it is not the sync signal, the process returns to step S52 again. On the other hand, when the sync signal is detected, the process proceeds to step S5A and waits for the end of the sync signal.

When the end of the sync signal is detected, the decrement timer for confirmation immediately before the frame starting point is preset to the initial value (frame starting point monitoring time-Δt) (step S5).
B). Here, Δt is a margin for preventing a misjudgment due to a subtle difference in the frame origin of each node. Then, in steps S5C and S5D, it is confirmed that the transmission path does not become "1" within this period. Furthermore, it is checked in a frame check process (step S5E) described later whether or not one frame of data can be normally received, and if normal, the process proceeds to the main loop.

If it is abnormal, the process returns to step S52 and tries to enter again.

Since the processing procedure in the main loop is the same as that of the first embodiment (see FIG. 4) except that the initial value of the sync signal output decrement timer (arbitrary time of 2 bit length or more) is different, it is omitted. .

Further, transmission / reception of each data block is as follows.
Rather than being separated by time, it is done with an arbitrary length specified by the header and trailer.

FIG. 18 is a timing chart in the fourth embodiment and is equivalent to FIG. As shown in the figure, each data block has the optimum length and the synchronization signal is short, so that the transmission efficiency can be improved as compared with the case of FIG.

The transmission efficiency can also be improved by applying it to the above-mentioned third embodiment in the same manner.

[0112]

According to the present invention, not only the polling sequence by the control station and the token transfer sequence can be eliminated, but the data itself flowing on the transmission line has the signal generation timing "identify the originating node". Information "or" information for identifying the type of data ", it is not necessary to include the" source address "or" type of data "in the data sent to the transmission line, and the transmission line can be used very efficiently. it can.

Moreover, if the accumulated time of no signal is used, even if a specific node does not communicate due to a failure or other reasons, or there is data that is not output depending on the type of data, the idle time of the transmission path Can be minimized, and the use efficiency can be further improved.

Further, since there is no collision between nodes in use of the transmission path and re-outputting due to the collision, the transmission delay is not always stable, and even if an arbitrary node stops functioning, the transmission delay between other nodes is increased. The communication can be stopped even temporarily, and the transmission delay can be continued without being larger than that in normal times, and it is possible to exert a very excellent effect on fault tolerance and real-time property.

[0115]

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a processing node according to the present invention.

FIG. 2 is a configuration example of a transmission line and an eliver according to the present invention.

FIG. 3 shows a signal format according to the present invention.

FIG. 4 is a flowchart showing a procedure of communication processing according to the present invention.

FIG. 5 is a flowchart showing a procedure of communication processing according to the present invention.

FIG. 6 is a flowchart showing a procedure of communication processing according to the present invention.

FIG. 7 is a flowchart showing a procedure of communication processing according to the present invention.

FIG. 8 is a timing chart showing the processing contents of each node and the state of the transmission path in the present invention.

FIG. 9 is a timing chart showing the processing contents of each node and the state of the transmission path in the present invention.

FIG. 10 is a timing chart showing the processing contents of each node and the state of the transmission path in another embodiment of the present invention.

FIG. 11 is a timing chart showing the processing contents of each node and the state of the transmission path in another embodiment of the present invention.

FIG. 12 is a flowchart showing a procedure of communication processing in another embodiment of the present invention.

FIG. 13 is a flowchart showing a procedure of communication processing in another embodiment of the present invention.

FIG. 14 is a timing chart showing the processing contents of each node and the state of the transmission line in another embodiment of the present invention.

FIG. 15 is a flowchart showing a procedure of communication processing in another embodiment of the present invention.

FIG. 16 is a flowchart showing a procedure of communication processing in another embodiment of the present invention.

FIG. 17 is a flowchart showing a procedure of communication processing according to the present invention.

FIG. 18 is a timing chart showing the processing contents of each node and the state of the transmission line in another embodiment of the present invention.

FIG. 19 is a diagram showing an overall configuration of a distributed control system.

FIG. 20 is a timing chart showing the processing contents of each node and the state of the transmission path in the related art.

[Explanation of symbols]

 B data transmission path # 1 to # 4 processing nodes 11 to 16 controlled equipment 20 communication interface 30 communication processing device 31 transmission processing means 32 reception processing means 33 synchronization signal output means 34 frame start point detection means 35 transmission path use period determination means 40 Control processor 50 I / O port

Claims (5)

[Claims] 1. A plurality of processing nodes are connected to a bus-type data transmission line, and each processing node is configured by a group of devices connected to each processing node by exchanging information with each other via the data transmission line. In the distributed control system for controlling the entire system, each of the processing nodes outputs a synchronization signal for autonomous communication synchronization to the data transmission path, and the data transmission path. Frame start point detecting means for detecting the starting point of the next transmission frame by monitoring the elapsed time from the time when all the synchronization signals of the respective nodes become non-signals; And a transmission path use period determining means for autonomously determining the transmission path use period of the distributed control data transmission system. 2. The distributed control according to claim 1, wherein the transmission path use period of the own node is determined by comparing the elapsed time from the transmission frame starting point with a value given in advance for each node. Data transmission system. 3. The transmission path use period of its own node is determined by comparing the accumulated time of no signal state in the data transmission path after the transmission frame starting point with a value given in advance for each node. The data transmission system for distributed control according to claim 1. 4. The transmission path use period of the self node is determined by comparing the elapsed time from the transmission frame starting point with a value given in advance for each type of output data. Data transmission system for distributed control. 5. The transmission path use period of its own node is determined by comparing the accumulated time of no signal state in the data transmission path after the transmission frame starting point with a value given in advance for each type of output data. Claim 1 characterized by the above-mentioned.
The data transmission system for distributed control described.