WO2013121568A1

WO2013121568A1 - Node synchronization method, network transmission system, and node device

Info

Publication number: WO2013121568A1
Application number: PCT/JP2012/053724
Authority: WO
Inventors: 幸輝湯尾
Original assignee: 富士電機株式会社
Priority date: 2012-02-16
Filing date: 2012-02-16
Publication date: 2013-08-22

Abstract

This slave node receives a synchronization frame transmitted by this master node, thereby initializing a timer of the slave node. The slave node then transmits a transmission received frame to the master node. The master node receives the transmission received frame transmitted by the slave node, and from the difference between the timer time at reception and the timer time when the synchronization frame that was transmitted from the slave node, the master node calculates the round-trip transmission delay time between the master node and the slave node, and transmits a round-trip transmission delay time notification frame notifying the slave node of the calculated round-trip transmission delay time. The slave node receives the round-trip transmission delay time notification frame, and on the basis of the round-trip transmission delay time included in the notification frame, synchronizes the slave node timer with the master node timer.

Description

Node synchronization method, network transmission system, and node device

The present invention relates to a node synchronization method, a network transmission system, and a node device for realizing stabilization of a data exchange cycle of each node.

For example, in an industrial network such as a conventional transmission system for plant control, it is necessary for each device constituting the system to exchange large amounts of data with each other while guaranteeing real-time performance. Therefore, for example, when mutual access is performed in an event according to the occurrence of an access request by an application installed in each device, the network load depends on the application, and real-time performance cannot be guaranteed.

Therefore, conventionally, there is a technology in which each device has a virtual shared memory (common memory) and transmits its own station (own node) data to all stations on the network at the update timing. When such a technique is used, each received station updates its data and accesses an application, thereby realizing a data exchange system that guarantees real-time performance.

Also, conventionally, a method for realizing efficient broadcast communication (broadcast communication) on the network at the time of data exchange described above has been proposed (see, for example, Patent Document 1). In Patent Document 1, the time division multiple access method using the internal timer of each node and the internal timer correction of the slave node using the synchronization frame from the master node are used together. In the method shown in Patent Document 1, the transmission path is configured as a network connected by a bus or a serial cable.

JP 2005-159754 A

Recently, Ethernet (registered trademark) has been applied to industrial networks, and it is becoming mainstream in the network at the controller level considering cooperation with information system devices. Therefore, when the technique as disclosed in Patent Document 1 is applied to a line using Ethernet as a physical layer, there are the following problems.

For example, in the technique shown in Patent Document 1, the physical layer is a cascade of buses or serial cables. Therefore, it can be assumed that the reception timing of the data transmitted by broadcast is a time difference that can be ignored at the same time or almost at each station.

However, for example, when Ethernet such as 100BASE-TX or 1000BASE-T is used as a transmission path, a star type is adopted as a topology, and a configuration including a relay device such as a HUB (hub) is adopted. In general, HUB employs a relay method called store & forward when relaying frame data, and transmission to the destination node is not performed unless all the frame data is received. For this reason, the reception timing delay at each node of the broadcast transmission data varies depending on the number of HUB stages that pass through. Therefore, for example, when a high-efficiency data exchange is realized using a high-speed communication line such as 1000BASE-T, the delay as described above cannot be ignored.

Further, for example, when the reception timing of the synchronization frame transmitted by the master station varies, the timer correction timing of each slave station varies, resulting in duplication (congestion) of data transmission timing in each station. . Congestion is fatal for an industrial network because frame data is discarded due to buffer capacity such as HUB as well as data delay. In particular, when congestion occurs in the synchronization frame of the master station, the reception timing of the synchronization frame is not only delayed even if the frame is not discarded, but also delayed depending on the size and type of the congested frame. Time is indeterminate. Therefore, the conventional method has a problem that the data exchange cycle in the entire system is disturbed.

The present invention has been made in view of the above points, and an object thereof is to provide a node synchronization method, a network transmission system, and a node device for realizing stabilization of a data exchange cycle of each node.

In order to solve the above problems, the present invention provides a node synchronization method in which a master node having a timer and a plurality of slave nodes are connected via a communication path, and the timers of the respective nodes are synchronized. By receiving the synchronization frame transmitted by the master node, the slave node initializes the timer of the slave node and transmits a reception completion frame to the master node. The master node transmits by the slave node. Round-trip transmission between the master node and the slave node from the difference between the time of the timer at the time of reception and the timer time at the time of transmission of the synchronization frame transmitted from the own node The delay time is calculated, and the calculated round trip transmission delay time is notified to the slave node. The slave node receives the round-trip transmission delay time notification frame, and sets the timer in the own node based on the round-trip transmission delay time included in the notification frame. Synchronize with.

In addition, what applied the arbitrary combination of the component of this invention, the expression, or the component to a method, an apparatus, a system, a computer program, a recording medium, a data structure, etc. is effective as an aspect of this invention.

According to the present invention, stabilization of the data exchange cycle of each node can be realized.

It is a figure which shows an example of schematic structure of the network transmission system in this embodiment. It is a figure which shows the example of transmission of the data before node synchronous correction | amendment. It is a figure which shows an example of the data exchange procedure for slave node time correction | amendment. It is a figure which shows the example of transmission of the data after node synchronous correction | amendment. It is a figure which shows the example of a system configuration for demonstrating how to determine margin time. It is a flowchart which shows an example of the master determination in a node synchronous process. It is a flowchart which shows an example of the slave determination in a node synchronous process. 10 is a flowchart illustrating an example of switching from a slave to a master in node synchronization processing. It is a figure which shows an example of a function structure of the node apparatus in this embodiment. It is a figure which shows an example of the hardware constitutions of a node apparatus.

<About the present invention>
In the present invention, for example, a timer of a node (station) configuring a network is adjusted not to a synchronization frame reception timing but to a synchronization frame transmission timing of a master station. In other words, the present invention performs synchronization matching that minimizes the influence (for example, synchronization error) due to the relay delay of a frame such as a relay device (for example, HUB) inserted into a communication path (network), for example. Perform communication cycle synchronization with high accuracy. Here, the synchronization in the present embodiment refers to a state in which, for example, the phase (timer) of the counter operation of each node is matched, but is not limited to this. In addition, the present invention can perform communication without causing data congestion on the network at the time of transmission of the synchronization frame by having a margin before transmission of the synchronization frame.

Hereinafter, embodiments in which a node synchronization method, a network transmission system, and a node device according to the present invention are suitably implemented will be described with reference to the drawings.

<Network transmission system: schematic configuration example>
FIG. 1 is a diagram illustrating an example of a schematic configuration of a network transmission system according to the present embodiment. As an example, the network transmission system 10 shown in FIG. 1 includes a plurality of node devices 11 (node devices 11-1 to 11-3 in the example of FIG. 1) and a HUB (hub) 12 as one or a plurality of relay devices. (HUBs 12-1 to 12-5 in the example of FIG. 1). Note that the number and types of node devices and relay devices, and connection methods are not limited to this.

In the example of FIG. 1, the node device 11-1 as the node A is a master node (master station), and the node device 11-2 as the node B and the node device 11-3 as the node C are slave nodes ( Slave station). As shown in the example of FIG. 1, the communication path of the network transmission system 10 is, for example, a star type having a relay device between the master node 11-1 and the slave node 11-2. The relay apparatus uses HUB as an example, but the present invention is not limited to this, and for example, a router, a repeater, an optical converter, or the like can be used.

In this embodiment, the master node and the slave node are connected via a communication path. In addition, since the master node and the slave node are used as, for example, a network of controllers (control devices), basically, devices used in the controller system are targets of masters or slaves. Examples of the device include a controller, a PC (Personal Computer), a server, an input / output (IO) module, and a drive device (for example, an inverter and a servo).

In the network transmission system 10 shown in FIG. 1, the node device 11-1 and the node device 11-3 are connected to the same HUB 12-1, and the node device 11-2 passes through a 5-stage HUB (relay device). The node device 11-1 and the node device 11-3 are connected.

Here, in order to clarify the contents of the node synchronization method in this embodiment, the contents before and after the node synchronization correction will be described. FIG. 2 is a diagram illustrating an example of data transmission before node synchronization correction. In the example of FIG. 2, an example of data transmission when the node device 11-2 (node B) or the connected HUB 11-2 is powered on and the node device 11-2 newly joins a communication path (network). Is shown. Note that the node device 11-1 (node A) and the node device 11-3 (node C) at this time are already synchronized.

In the example of FIG. 2, each node has a cycle timer (first timer) and a send timer (second timer). In the example of FIG. 2, the synchronization frame 21 (in the example of FIG. 2, the synchronization frames 21-1 to 21-4) is transmitted according to the cycle T of the cycle timer (A) of the node device 11-1. Yes.

Here, the node device 11-1 also clears the send timer 22 (send timers 22-1 to 22-3 in the example of FIG. 2) at the cycle timer clear timing. Thereafter, the node device 11-1 transmits the transmission data 23 (transmission data 23-1 to 23-3 in the example of FIG. 2) of each node at the timing of the send timer up.

At this time, the node device 11-2 receives the synchronization frame 21 from the node device 11-1 via the HUB 12-1. Therefore, a delay 24 occurs at the reception timing of the synchronization frame 21. At this time, since the node device 11-2 (node B) is not synchronized, the cycle T and the send timer (B) of the node device 11-2 are cleared at the reception timing of the synchronization frame 21. Done.

Therefore, the send timer (B) operates in a state including the delay 24. Strictly speaking, the node device 11-3 (node C) also includes a delay corresponding to one HUB, but here it is not considered because it is a little in time.

Next, in the example of FIG. 2, data 25 (data 25-1 to 25-3 in the example of FIG. 2) is transmitted at the timing of the send timer up of the node device 11-2 in the state described above. In this case, the node device 11-1 and the node device 11-3 overlap with, for example, the synchronization frames 21-2 to 21-4 at the timing of receiving the data 25, and congestion occurs. In this embodiment, control can be performed so that data transmission from a participating node (for example, the node device 11-1) is not performed until synchronization is completed. In order to explain, an example expressing congestion is shown.

Here, a general Ethernet HUB employs an interface method called store & forward. In this case, all the sent frames are stored in the reception buffer in the HUB, and are transmitted after performing the HUB internal processing (for example, abnormality determination, destination determination, etc.). In addition, it is normal that the reception buffer assuming that the destination port is in use has a sufficient size. If there is some congestion, communication is performed without data loss. be able to. However, the reception timing of the synchronization frame is an important element for synchronization, and delayed reception of the synchronization frame is fatal.

Further, for example, in Ethernet, frame priority control can be performed by QoS (Quality of Service) or the like. However, in the example of FIG. 2, since the data 25 from the node device 11-2 has already been received, the synchronization frame 21 is also delayed in reception by the node device 11-3 regardless of the priority setting. The synchronization of each node on the system will be lost.

HUB used in Ethernet does not have an IP address (Internet Protocol Address) and is not normally recognized by the system configuration node. In addition, since a HUB can be constructed in a star topology, it is possible to construct a system via a number of stages of HUB between system configuration nodes, and the delay time due to the HUB varies depending on the manufacturer, model, etc. There are many cases. Therefore, in practice, it has been difficult to set a delay time from the master node to each slave node until the system is configured.

Therefore, in this embodiment, the transmission delay time between the master node and the joining slave node is determined, and this is set as the joining slave node, so that the master timer and the joining slave timer are synchronized, and then joined to the system. Can be made.

Here, FIG. 3 is a diagram showing an example of a data exchange procedure for correcting the slave node time. In FIG. 3,

TC bands

31 and 37 indicate the transmission band (timer clear) of the synchronization frame, and

TS bands

32 and 38 indicate the data transmission band (timer send), and each is transmitted at a predetermined cycle. Is done.

Further, the nodes A, B, and C in the example of FIG. 3 correspond to the node devices 11-1, 11-2, and 11-3, respectively. Also, the squares shown in FIG. 3 indicate frames, the squares on the lines for the nodes A to C indicate transmission frames, and the squares below the lines indicate reception frames.

Also, the TC frame indicates a synchronization frame, and the TS * (* indicates any of A to C) frames indicates node data from any of the nodes A to C. For example, TSA indicates that the node data is from node A (node device 11-1). In FIG. 3, a req frame 33 indicates subscription request data, an OK frame 34 indicates subscription permission data, a dly frame 35 indicates a TC reception delay, and a SET frame 36 indicates a delay time setting (specifically, Indicates round trip transmission delay time setting).

Here, the node B (node device 11-2) that is the joining node transmits the subscription request req frame 33 in the TS band 38 to the node A (node device 11-1) that is the master node. The node A determines whether or not the participation requesting node can participate, and if the participation is permitted, the node A issues an OK frame 34 for permission to join following the synchronization frame TC. In the present embodiment, the OK frame 34 is expressed as a separate frame. However, the present invention is not limited to this. For example, the OK frame 34 may be configured as data on the synchronization frame TC.

The node B (node device 11-2) as a slave node that has received the OK frame 34 that has been permitted to join from the node A transmits a dly frame 35 that is a TC reception delay frame to the master at the next synchronization frame TC reception timing. To do. This frame is transmitted at the timing of receiving the synchronization frame TC. However, the node device B, which is a joining node, has already received the synchronization frame TC several times, and has cleared (initialized) the internal timer at that timing. For this reason, the node B, which is a joining node, may transmit at the clear timing of the cycle timer of the own node (own station) in the same manner as the synchronization frame TC transmission of the master.

The dly frame 35, which is a TC reception delay frame, is unicast (one-to-one) communication to the master node, and thus overlaps the reception of the synchronization frame TC to the slave node that has already joined the network transmission system. There is nothing.

Also, the node A that has received the dly frame 35 records the value of the cycle timer at that timing. The reception timing of the dly frame at the node A is considered to be equivalent to the response of the joining slave to the synchronization frame TC transmitted by the node A. Therefore, the difference between the cycle timer value of the reception timing of the dly frame 35 and the cycle timer value (= 0) at the time of transmission of the synchronization frame TC is considered as the round trip transmission time (round trip transmission delay time) between the master and the joining node. It is done.

Therefore, the node A sets the transmission time obtained above in the SET frame 36 for setting the delay time for the joining slave node (node B), and uses this SET frame 36 as a round trip transmission delay time notification frame. Send to. The Node B that has received the SET frame 36 subtracts the round-trip transmission delay time set in the SET frame 36 from its own cycle timer (accelerates timer up) (for example, time 39 shown in FIG. 3), The cycle timer is synchronized with the timer of the node A that is the master node. In the example of FIG. 3, the cycle timer of the node B is delayed from the timer of the node A, and thus the synchronization is performed by the above-described subtraction process. However, the present invention is not limited to this. For example, when the cycle timer of the node B is ahead of the timer of the node A, the round-trip transmission delay time set in the SET frame 36 is added to the own cycle timer so that the own cycle timer is set at the master node. It can be synchronized with the timer of a certain node A.

In the present embodiment, in FIG. 3, the delay time setting SET frame 36 may also be data of the synchronization frame TC, similar to the subscription-permitted OK frame 34. In this way, the node that has finished adjusting the cycle timer is allowed to transmit data in the TS band, assuming that it is permitted to join the system.

In the present embodiment, the req frame 33 is used as the subscription request data and the OK frame 34 is used as the subscription permission data frame for the node subscription. However, the present invention is not limited to this. The synchronization in this embodiment can be performed even in a procedure in which station numbers are sequentially designated in the framing frame TC and a station scheduled to join among the designated stations (nodes) responds with the dly frame 35.

<About extra time>
Here, even when the synchronization is performed as described above, the node that is farther in transmission time than the node that finally transmits data is congested with the data of the slave node that was transmitted last and the synchronization frame TC of the master node. May be received. Note that a node far in terms of transmission time is, for example, a node having a predetermined number or more of the number of HUB stages passed through.

Therefore, in this embodiment, a margin time can be provided between the timing of the final data transmission and the transmission timing of the synchronization frame TC. However, the conventional method depends on the system configuration and is not uniquely determined by the number of nodes or the like.

On the other hand, since the master node (node device 11-1) in the present embodiment performs the above-described joining process for all slave nodes of the network transmission system 10, the transmission delay time between the master and each slave is maintained. Yes. Therefore, although the master node is not necessarily at the center of the configuration, for example, a predetermined value based on the maximum transmission delay time (round-trip transmission delay time) for all slave nodes can be set as the margin time.

The margin time in the present embodiment is a time during which data transmission or the like provided for avoiding congestion, collision, etc. of data flowing through the network is not performed. In the present embodiment, it is possible to prevent the synchronization frame TC from being congested by providing a margin time as described above.

In the present embodiment, theoretically, if there is an allowance time for the maximum round-trip transmission delay time, there is no overlap with the next synchronized frame transmission timing of the master node. However, considering the margin and efficiency in actual operation, the margin time may be longer than the maximum round trip transmission delay time, for example, preferably about 2 to 3 times.

Here, FIG. 4 is a diagram showing an example of data transmission after node synchronization correction. In the example of FIG. 4, the margin time 41 is set as described above. In the present embodiment, data transmission between nodes can be completed by setting a margin time 41 as shown in FIG. That is, in the present embodiment, a value that allows for the margin time 41 is set for the cycle T of the cycle timer.

That is, in the present embodiment, when synchronization is performed between the master node and the slave node, the slave node sequentially responds simultaneously with a preset send timer up after the master node transmits a synchronization frame. At this time, the margin time 41 means, for example, a time until the next synchronization frame transmission of the master node with respect to the transmission timing of the last slave node. That is, the margin time 41 means the time from the timing at which all the slave nodes have transmitted responses to the synchronization frame transmission of the master node to the timing at which the next synchronization frame is transmitted.

Thus, in this embodiment, it is ensured that no data communication is performed in the communication band of the synchronization frame TC, and that the synchronization frame TC is received with a constant delay time for each slave node. .

When there is no data congestion, the frame processing time in the HUB 12 is constant. Therefore, as long as the system configuration does not change, frames can always be received with the same delay time.

Here, the above-described method for determining the allowance time (how to calculate) will be specifically described. FIG. 5 is a diagram illustrating an example of a system configuration for explaining how to determine the allowance time.

Data congestion or collision occurs when data being communicated is delayed, but in a network transmission system, most of the delay elements are relay devices such as HUBs.

Therefore, for example, the margin time is set in accordance with the number of data relay apparatuses in the system configuration. Specifically, for example, the data arrival time of the route having the largest number of data relay apparatuses among the communication between nodes can be used, but the present invention is not limited to this.

For example, in the network transmission system 50 shown in FIG. 5, eight node devices 51-1 to 51-8 of the nodes A to H and seven HUBs 52 that are relay devices that relay the nodes A to H in multiple stages. -1 to 51-7. Note that the number and types of node devices and relay devices, and connection methods are not limited to this.

Here, in the present embodiment, the master node is not determined in advance, but is determined by, for example, a master determination method in node synchronization processing of the present embodiment described later. Therefore, all of the nodes A to H can be master nodes unless the node device 51 has a special setting (such as a slave mode).

In addition, when the node A or the node H becomes the master node, the number of relay stages of the HUB 52 is the largest between the node A and the node H (7 stages in the example of FIG. 5). Therefore, in this embodiment, the system can be operated without problems such as data congestion and collision by setting a margin time based on, for example, the data arrival time from the node A to the node H, the maximum delay time, and the like.

On the other hand, for example, when the node D or node E becomes the master node, the maximum delay time detected is the number of relay stages from the node D or node E to the node A or node H (four stages in the example of FIG. 5). It turns out that. Further, the detected maximum delay time is minimum when the central node of the system configuration becomes the master. In the example of the network transmission system 50 in FIG. 5, the maximum delay time is minimum when the node D or the node E is the master node. Become. In other words, if the master node is at the center of the system configuration, the maximum delay time in that configuration is between the slaves arranged opposite to each other across the master node (in the example of FIG. 5, node A, node H).

Therefore, in this embodiment, when setting the margin time, it is possible to include the delay time between all nodes by setting about 2 to 3 times the maximum delay time from the master node as a reference. It becomes possible. Note that the margin time increases as the master node deviates from the center of the system configuration.

In this embodiment, for example, when a node that joins the system later is farthest from the master, the calculation of the margin time may change. However, in a system that requires strict synchronization as in the present embodiment, it is difficult to think about joining a node to the system that is not initially assumed. For this reason, in this embodiment, for example, dropping of a node or additional joining is not permitted, or if the joining node is farthest from the master, the system is abnormally terminated. In the present embodiment, the “maximum cascade stage number × 2 times” of the relay device can be uniquely determined as a margin time. In the present embodiment, the maximum number of stages of the relay device of the system can be set in advance by the user, and abnormal termination or the like can be performed when the number of stages is exceeded.

<Slave exceeding the allowable value>
Here, in the network transmission system according to the present embodiment, for example, a node exceeding a preset margin time is prevented from joining the system. For example, if the master node cannot receive within the TC band when receiving a subscription request, it is possible to control the subscription to the system by determining that a node exceeding the margin time has been inserted.

In other words, even if the master node is a slave node that has requested to join, the master node refuses to join a node that exceeds the above-described margin time, and the slave node notifies the user that there is an abnormality in joining. When rejecting the subscription, for example, processing such as setting no delay time or responding abnormally is performed.

<About resetting synchronization>
Next, the resetting of synchronization in this embodiment will be described. The resetting of synchronization means, for example, recalculation or resetting of delay when there is a configuration change. The delay time is set, for example, when the node system is joined or when the master is disconnected. In the present embodiment, the delay time may be measured a plurality of times (several tens of times) when the system joins or the master leaves, and the average value may be set as the delay time between nodes. However, in an actual system, the state may change due to aging of nodes and devices, ambient environment (for example, temperature, noise), and the like.

Therefore, in this embodiment, delay time measurement is performed for all nodes participating in the network transmission system, for example, at a constant cycle (for example, once every several seconds), and the delay time is set again.

<Method for detecting and correcting synchronization loss during system operation>
Next, a method for detecting and correcting synchronization loss during system operation will be specifically described. During system operation, the synchronization is basically adjusted on the slave side. For example, in the present embodiment, a delay time is distributed from the master node to each slave node, and the timer is adjusted correspondingly. While the slave node participating in the system can synchronize with the master node, the synchronization frame TC transmitted from the master node is received with a delay of the delay time. That is, in this embodiment, the reception timing of the synchronization frame TC is measured by the timer of the own node (own station), and the time is compared with a preset delay time at the time of synchronization with the master node. It is possible to detect whether or not synchronization has occurred.

Thereby, in the present embodiment, for example, when the TC reception timing is confirmed a plurality of times, and reception of TC is delayed from a set delay time in advance by a predetermined number (for example, 5 times) in advance, etc. Based on the set condition, the delay or advance of its own timer is detected. Further, in the present embodiment, based on the amount of delay or advance of own timer obtained at the time of detection described above, a correction for synchronizing the timer in the own node with the timer of the master node so as to eliminate the above-described synchronization deviation. I do.

<Master determination method in node synchronization processing>
Next, the node synchronization processing in the present embodiment will be specifically described using a flowchart. FIG. 6 is a flowchart illustrating an example of master determination in node synchronization processing. 6 illustrates a procedure in which one of a plurality of node devices connected to the network is determined and operated as a master node, but is not limited thereto.

In the process of FIG. 6, the node synchronization process first initializes each resource of the node device (S01), and monitors the initial line when it is normally initialized (S02). Here, monitoring of the initial line means monitoring of the line during initial (initialization), for example, and is distinguished from monitoring during operation. For example, during initialization, frames such as TC (synchronization frame transmission band (timer clear)), TS (data transmission band (timer send)), and ITC (TC frame used only during system initialization) are included. There is a possibility of flowing. On the other hand, the ITC does not flow during operation, unlike during initial operation. In the process of S02, for example, a line status check is performed. The check cycle can be checked at a timing such as “maximum cycle time × 3 times”, but is not limited to this, and may be checked at another timing.

The node synchronization process determines whether the monitoring timer is up (UP) without receiving the above-described TS, ITC, or TC as a result of monitoring (S03).

Here, in the node synchronization process, when the condition of S03 described above is satisfied (YES in S03), an initial temporary master is determined (S04). The initial temporary master monitors, for example, a station number (node number), a refresh cycle delay, TC reception of other node devices (other stations), and the like. The initial temporary master controls ITC transmission. Next, in the node synchronization processing, for example, it is determined whether or not the ITC transmission is equal to or greater than a predetermined number (for example, three times) and there is a slave joining notification (S05), and the ITC transmission is determined to be a predetermined number (for example, 3). If there is a slave subscription notification (YES in S05), the master right is acquired and the master operation is performed (S06). Here, the master operation in the present embodiment is, for example, at least one of the above-described TS reception monitoring, joining node management, transmission delay time calculation, message transmission control, delay time setting transmission for each slave node, and the like. However, the present invention is not limited to this.

In the node synchronization process, if the above-described condition of S05 is not satisfied (NO in S05), it is determined whether another node apparatus TC is received or an ITC having a higher priority than the own node is received. Judgment is made (S07). In the node synchronization process, when another node apparatus TC is received or an ITC having a higher priority than the own node is received (YES in S07), a process for slave operation described later is performed (S08).

In the node synchronization process, when the condition of S03 is not satisfied within the predetermined time in the process of S03 (NO in S03), or when the condition of S07 is not satisfied within the predetermined time in the process of S07 (S07). NO), the node synchronization processing is terminated.

Further, in the present embodiment, in each process described above, the process described above is repeated when there is a mild abnormality that can be easily recovered, such as a process delay in memory read / write or a minor communication failure. Done. In this embodiment, for example, when there is a serious abnormality that cannot be easily recovered, such as a failure, a process of notifying error information may be included.

<Slave determination method in node synchronization processing>
Next, FIG. 7 is a flowchart showing an example of slave determination in the node synchronization processing. In the example of FIG. 7, a procedure is shown in which one of a plurality of node devices connected to the network is determined and operated as a slave node, but is not limited thereto.

In the process of FIG. 7, the node synchronization process first initializes each resource of the node device (S11), and monitors the initial line when it is normally initialized (S12). In the process of S12, for example, the line status is checked. The check cycle can be checked at a timing such as “maximum cycle time × 3 times”, but is not limited thereto.

The node synchronization processing determines whether there is reception of TS, ITC, or TC as a result of monitoring (S13). Here, in the node synchronization process, when the condition of S13 described above is satisfied (YES in S13), an initial master is determined (S14).

Note that the initial master determination in the process of S14 includes, for example, “a state in which there is a master (or master candidate) other than the own node and the own node may become a slave”. Basically, when the own node is in the line monitoring state after initialization and a frame such as TC has already flowed, the own node may be a slave. However, in an unstable state such as when the system is started up, there is a risk that the master is determined by receiving a single frame. Therefore, in this embodiment, monitoring is performed for a predetermined period (for example, three periods), and if that is all right, the node that transmits TC (or ITC) is recognized as a master, and the own node tries to become a slave. To do.

Specifically, the initial master monitors, for example, TC, TS, etc., receives ITC or TC, and checks the master station number. This monitoring is performed with reference to, for example, three laps, but is not limited to this.

Next, in the node synchronization processing, it is determined whether or not a TC or ITC of the same node has been received for a predetermined number of consecutive times (for example, three consecutive) (S15), and a TC or ITC of the same node has been received for a predetermined number of consecutive times If this is the case (YES in S15), a slave operation confirmation wait state is entered (S16). At this time, in this embodiment, the network cannot be joined. In the process of S16, for example, a TS transmission delay time is set and a delay time backup for the connection node is performed.

Further, the node synchronization processing determines whether or not the master is permitted to join and the delay time setting is completed (S17). If the master is permitted to join and the delay time setting is completed (YES in S17). Then, slave operation is performed (S18).

In the node synchronization process, when the condition of S13 is not satisfied within the predetermined time in the process of S13 (NO in S13), or when the condition of S15 is not satisfied within the predetermined time in the process of S15 (S15 NO), the node synchronization processing is terminated. In the present embodiment, in the process of S15 described above, when the condition of S15 is not satisfied within a predetermined time, the own node may try to become the master. In this case, after the slave operation is confirmed, if the master is not allowed to join, an error notification may be given due to network setting abnormality or the like.

Further, in the node synchronization process, when the condition of S17 is not satisfied within the predetermined time in the process of S17 (NO in S17), the node synchronization process is terminated.

<Switching method from slave to master in node synchronization processing>
Here, in this embodiment, when an abnormality such as a failure occurs during the master operation, it is preferable to switch the node that has been operated as a slave to the master.

FIG. 8 is a flowchart showing an example of switching from the slave to the master in the node synchronization processing. In the example of FIG. 8, it is assumed that slave operation is performed as described above in the node synchronization processing (S21). In such a state, when it is detected that the master is lost due to a failure of the master node or the like (S22), the node synchronization processing makes an inquiry about master right transfer on the network (S23). Specifically, the TS with the inquiry bit turned ON is transmitted (S24). In the above-described inquiry, for example, when an ITC having a higher priority than the own node is received or there is an inquiry bit in a TS frame having a higher priority than the own node, the node synchronization processing is performed for a predetermined time. Wait for operation master switching.

Also, the node synchronization processing is set as an operational temporary master when it is detected that the master has been dropped in the operation master switching waiting state in the processing of S22 described above (S25). At this time, the node synchronization processing performs ITC transmission at a predetermined timing. Further, the node synchronization processing waits for switching of the operation master when ITC or TC having a higher priority than the own node is received.

Next, in the node synchronization process, it is determined whether or not the ITC has been transmitted continuously for a predetermined number of times (for example, three times) within a predetermined time (S26), and when the predetermined number has been transmitted continuously (YES in S26). The master operation is switched (S27). Further, in the node synchronization process, when a predetermined number of ITCs cannot be transmitted continuously within a predetermined time (NO in S26), the process is terminated and the operation master is waited for switching.

In other words, the participating nodes of the network in operation always monitor the frame, the master node monitors the TC frame from the master having higher priority than itself, and the slave node monitors the drop of the master node. . The slave node that detects the drop of the master node adds a master transfer inquiry to the TS frame, and performs a master transfer process.

In this embodiment, when both the master and the slave detect a TC frame from a station having a higher priority than the node that is currently the master node, the master switching state is entered, and the normal operation is entered after the master is determined. can do. The priority described above is determined by, for example, the type of node (for example, module shape, PCI-e (Peripheral Components Interconnect-express) board), the network station number (node number) in ascending order, and the like.

<Example of functional configuration of node>
Next, a functional configuration example of the node in the present embodiment will be described with reference to the drawings. FIG. 9 is a diagram illustrating an example of a functional configuration of the node device according to the present embodiment. The node device 60 shown in FIG. 9 includes a communication interface (IF) means 61, a load command means 62, a storage management means 63, a system management means 64, a switch (SW) management means 65, and a time management means (timer). ) 66, application control means 67, network control means 68, and input / output (IO) control means 69.

The node device 60 shown in FIG. 9 has a common configuration for both the master node and the slave node. Also, the node device 60 shown in FIG. 9 has a configuration including both functions of the controller and the network. Here, the processing of the controller is performed, for example, with a constant cycle as a control unit (scan cycle). In this embodiment, an interrupt with a constant cycle from a control LSI (Large Scale Integration) (for example, a master LSI) is performed. Can be operated on the basis. In the present embodiment, the present invention is not limited to this. For example, processing may be performed based on a built-in timer of a microcomputer.

In this embodiment, the reception interrupt of the inter-controller network is used, the delay time is calculated from the difference between the TC reception timing and the scan cycle interrupt from the master LSI, and correction is performed. The controller can operate by downloading a support program (loader) on the PC, a combination program, and an operation definition.

The node device 60 as a normal controller and a loader is connected to an external device or the like by a communication IF means 61 via, for example, a serial cable or a USB (Universal Serial Bus). Here, the transmitted / received communication is output from the communication IF unit 61 to the loader command unit 62.

The loader command means 62 exchanges corresponding processing by a protocol called a loader command. Specifically, loader commands include commands such as application start / stop (system management control), application data monitoring and modification (application management control), program, definition download (storage management control), etc. Acts on function.

The storage management means 63 is rewritable and stores various information in a flash memory such as an SD memory card, which is a non-volatile semiconductor memory that does not lose data even when the power is turned off. Management such as reading. The storage management unit 63 outputs the acquired information to at least one of the system management unit 64, the application control unit 67, and the network control unit 68 as necessary.

The system management unit 64 executes preset processing (for example, execution cycle, memory allocation, IO configuration management, initialization of each resource, etc.), status monitoring (failure diagnosis) of the node device 60, IO data exchange processing Etc. The system management means 64 performs processing as described above based on various data obtained from the storage management means 63, loader command means 62, SW monitoring means 65, time management means 66, application control means 67, and IO control means 69. Execute.

Specifically, the system management means 64 receives the synchronization frame transmitted after the master node timer is up when the network control means 68 is set as a slave node. Further, the system management means 64 initializes the above-described timer for the TC and TS transmission periods in correspondence with the received synchronization frame, and transmits the reception completion frame from the communication IF means 61 to the master node after the initialization.

Further, the system management unit 64 receives the transmission delay time notification frame from the master node when the network control unit 68 sets the slave node, and based on the received transmission delay time, Adjust. Specifically, when the timer in its own node is delayed from the transmission delay time of the master node, the system management means 64, for example, uses the received transmission delay time (specifically, the round-trip transmission delay time) as its own. By subtracting from the current time of the timer of the node, the node device 60 can synchronize the current time of the local node timer with the master timer. In addition, when the timer in the own node is ahead of the transmission delay time of the master node, the system management means 64 adds the above-described transmission delay time to the current time of the own node timer, so that the master timer Synchronize with.

Further, the system management unit 64 receives the synchronization frame reception completion frame from the slave node when the network control unit 68 sets the master node. Further, the system management unit 64 records the current time obtained from the RTC in the above-described timer of the TC and TS transmission cycle in the time management unit 66 based on the received synchronization frame reception completion frame.

Further, the system management means 64 determines the round-trip transmission delay between the master node and the slave node from the difference between the current time of the timer when receiving the reception completion frame and the timer time when transmitting the synchronization frame transmitted from the own node. Calculate time. Further, the system management means 64 notifies the calculated round trip transmission delay time to the slave node.

Furthermore, the system management means 64 sets a margin time during which data transmission or the like is not performed at the master node with reference to the maximum transmission delay time between a plurality of slave nodes. Further, the system management means 64 uses the interval from the slave node data transmission to the master node synchronization frame transmission as a margin time. As a result, in the present embodiment, it is possible to prevent communication congestion from occurring during transmission of the synchronization frame.

Further, the system management means 64 performs various settings, control, monitoring, and the like. The settings in the system management means 64 are basically read from the storage management means 63. The setting file writing for storage management is stored from the outside via the communication IF means 61 and the loader command means 62, for example.

The setting contents include, for example, system settings (for example, IO module configuration, control cycle settings, etc.), module settings (for example, module initialization operation settings, etc.), but are not limited thereto. is not. The system settings described above are settings for the entire system. For this reason, settings for individual modules are transmitted as messages through network control or IO control.

Further, the control in the system management means 64 performs operation control such as starting and stopping of applications, and other resets (initialization), for example. These operation controls are instructed from the loader command means 62 or the SW monitoring means 65, for example. Moreover, as control in the system management means 64, in addition to the content described above, for example, control of exchange processing of IO data during operation, network data, and the like is performed.

Further, the monitoring in the system management means 64 monitors module hardware failures, network, and system-wide abnormality information collected by network control at regular intervals. In addition, monitoring by the system management means 64 includes system shutdown, notification to the user (for example, notification using voice, light, etc.), notification to a predetermined computer (for example, management terminal) using a loader command ( For example, an error message or the like is performed.

The SW monitoring unit 65 monitors an external key switch (Key SW) (for example, power ON / OFF, etc.), a station number SW, and the like, and sends each SW signal obtained by, for example, a user operation to the system management unit 64. Output.

The time management means 66 acquires time information that serves as a reference for synchronization. Specifically, the date information acquisition unit 66 acquires a signal from, for example, an RTC (Real Time Clock), and manages time and the like by a calendar function or the like.
Here, the time management means 66 may have the cycle timer (first timer) and the send timer (second timer) described above. Further, the time management unit 66 may manage the above-described margin time obtained from the system management unit 64. The time management unit 66 outputs the acquired time information to the system management unit 64.

The application control means 67 stores various application programs for controlling applications such as the above-described various node synchronization processes implemented in the present embodiment. Control signals from the loader command means 62 and the system management means 64 are stored. Various application processes are executed based on the above.

The network control means 68 controls the entire network for sending and receiving messages and control data with other nodes, for example. Further, the network control unit 68 performs control for the own node to set to either the master node or the slave node based on the connection state with other node devices as described above.

Specifically, the network control means 68 performs control so as to reject the subscription when the slave node requesting to join is a slave node exceeding the margin time when the own node is the master node. Further, the network control unit 68 determines a temporary master from a plurality of node devices, operates the temporary master as a master node when the ITC transmission is a predetermined number or more and a slave join notification is received from another node device. Then, other node devices are operated as slave nodes.

The network control unit 68 executes predetermined processing when a node device is newly joined to a plurality of node devices connected to the communication path (communication network) or when the master node is dropped from the communication path. Re-execute the process to synchronize. Further, when the node is a slave node, the network control unit 68 measures the TC reception timing with the timer of the node, and compares the measured time with a preset delay time, thereby It is detected whether or not synchronization has occurred, and correction is performed so that the above-described synchronization is eliminated in accordance with the delay or advance of the timer obtained from the detection result.

The IO control means 69 controls the overall input / output of messages and control data with the connected external device.

In the node device 60 shown in FIG. 9, the network function is described on the same functional block. However, the present invention is not limited to this. For example, the IO control unit 69 is divided into separate modules. It can be implemented to exchange messages and control data between modules.

<Example of hardware configuration of node device>
Next, an example of hardware of the node device according to the present embodiment will be described with reference to the drawings. FIG. 10 is a diagram illustrating an example of a hardware configuration of the node device. In the hardware configuration example illustrated in FIG. 10, the node device 70 includes a basic system unit 71, an application control unit 72, and a network control unit 73.

The basic system unit 71 corresponds to the system management unit 64 described above, the application control unit 72 corresponds to the application control unit 67, and the network control unit 73 corresponds to the network control unit 68.

The basic system unit 71 includes system control means 81, SDRAM (Synchronous DRAM) 82, LAN (Local Area Network) 83, RTC 84, SD memory (Secure Digital memory) 85, and flash memory (Flash Memory) 86. And an internal bus management LSI 87, an SRAM (Static Random Access Memory) 88, and a master LSI 89. The application control unit 72 includes an application network unit 91, a master LSI 92, an SRAM 93, a flash memory 94, and a DDR-SDRAM (Double-Data-Rate Synchronous Random Access Memory) 95. Yes.

Here, in FIG. 10 described above, for example, the SX bus (high-speed bus) and the E-SX bus (super high-speed bus) are predetermined input / output (IO) control buses. The system control 81 performs control in the basic system of the node device 70. For example, the system control 81 controls transmission / reception of various data from an external device connected via the LAN 83, controls the operation of the entire system based on time information obtained from the RTC 84, and an application (program) such as the SD memory 85. Is read and executed. Further, the system control unit 81 may input / output various data from an external device via a USB or the like.

Master LSIs

89 and 92 are control LSIs for the IO control bus. DDR-SDRAM 95, SDRAM 82, and

SRAMs

88 and 93 are volatile memories, and

flash memories

86 and 94 are writable nonvolatile memories.

The internal bus management LSI 87 is an LSI (FPGA (Field-Programmable Gate Array) for connecting microcomputers. The internal bus management LSI 87 is connected to, for example, a shared memory (for example, a shared memory included in the master LSI 92 shown in FIG. 10). The internal bus management LSI 87 controls access to the SRAM of another CPU (Central Processing Unit) connected by the SX bus. CPU module such as PLC (Programmable Logic Controller), etc. It means a module that performs application computations, and the internal bus management LSI 87 handles interrupts between microcomputers. Do.

In the processor bus access described above, the own node (own station) area corresponds to the processor bus space in FIG. Note that the above-described processor means a module that executes, for example, a PLC application.

Also, the application network means 91 has a structure that mainly uses, for example, a dual-core microcomputer and processes each application and network. The hardware configuration described above is not limited to this in the present invention. For example, one microcomputer may be used.

Here, in the present embodiment, each function as shown in FIG. 9 can be realized by using the hardware configuration shown in FIG. Specifically, the node device 60 in this embodiment is set to either a master node or a slave node based on the connection state with other node devices. Further, when the node device 60 is set as a slave node, the node device 60 receives a synchronization frame transmitted by the master node, thereby initializing a timer of the slave node and transmitting a reception completion frame to the master node. Further, when the node device 60 is set as the master node, the node device 60 receives the reception completion frame transmitted by the slave node, and the timer at the time of transmission and the timer at the time of transmission of the synchronization frame transmitted from the own node. The round trip transmission delay time between the master node and the slave node is calculated from the difference with the time. Further, the node device 60 transmits a round trip transmission delay time notification frame for notifying the slave node of the calculated round trip transmission delay time, causes the slave node to receive the round trip transmission delay time notification frame, and is included in the notification frame. Based on the round-trip transmission delay time, the timer in the own node is synchronized with the timer of the master node.

In the present embodiment, a program (node synchronization program) for causing a computer to function as each unit included in the node device 60 described above is generated, and the generated program is installed in a computer such as a PC. Each node synchronization process described above can be realized.

As described above, according to the present invention, stabilization of the data exchange cycle of each node can be realized. In addition, according to the present invention, in a system having a star topology such as Ethernet, for example, a timer of each node is synchronized with a shared memory network using a time division multiplex transmission method, thereby improving transmission efficiency and data exchange. Efficiency and stabilization of the data exchange cycle can be realized.

In addition, this embodiment can be applied to a synchronization method when performing a series of operations in a large-scale facility such as a steel plant using a plurality of operations. It can be widely applied as a synchronization method.

The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to the specific embodiments, and various modifications, It can be changed.

10, 50 Network transmission system 11, 51 Node device 12, 52 HUB (relay device)
21 synchronization frame 22 send timer 23 transmission data 24 delay 25

data

31, 37

TC band

32, 38 TS band 33 req frame 34 OK frame 35 dly frame 36 SET frame 39 time 41

allowance time

60, 70 node device 61 communication interface (IF ) Means 62 load command means 63 storage management means 64 system management means 65 switch (SW) management means 66 time management means (timer)
67 Application Control Unit 68 Network Control Unit 69 Input / Output (IO) Control Unit 71 Basic System Unit 72 Application Control Unit 73 Network Control Unit 81 System Control Unit 82 SDRAM
83 LAN
84 RTC
85

SD memory

86, 94 Flash memory 87 Internal bus management LSI
88, 93 SRAM
89,92 Master LSI
91 Application network means 95 DDR-SDRAM

Claims

A node synchronization method in which a master node having a timer and a plurality of slave nodes are connected via a communication path, and the timers of the respective nodes are synchronized,
The slave node receives a synchronization frame transmitted by the master node, initializes a timer of the slave node, and transmits a reception completion frame to the master node.
The master node receives the reception completion frame transmitted by the slave node, and from the difference between the timer time at the time of reception and the timer time at the time of transmission of the synchronization frame transmitted from the own node, Calculate a round trip transmission delay time between the master node and the slave node, send a round trip transmission delay time notification frame to notify the slave node of the calculated round trip transmission delay time,
The slave node receives the round trip transmission delay time notification frame, and synchronizes a timer in the own node with a timer of the master node based on the round trip transmission delay time included in the notification frame. Node synchronization method to perform.
The master node is based on a maximum round-trip transmission delay time among the round-trip transmission delay times between a plurality of slave nodes.
The master node transmits the synchronization frame to all slave nodes, and sets a margin time from a timing at which all the slave nodes have transmitted a response to a timing at which the next synchronization frame is transmitted. The node synchronization method according to claim 1.
3. The node synchronization method according to claim 2, wherein the margin time is equal to or longer than the maximum round-trip transmission delay time.
The node newly requesting to join the communication path transmits a join request frame to the master node,
The master node, after determining whether or not to join the node requesting to join, when allowing participation as a slave node, transmits a join permission frame following the synchronization frame,
The node requesting to join receives the join permission frame and the synchronization frame, and as a slave node, transmits a reception delay frame to the master node at a reception timing of the next synchronization frame,
2. The master node according to claim 1, wherein a difference between a cycle timer value at the time of reception of the reception delay frame and a cycle timer value at the time of transmission of the synchronization frame is set as the round trip transmission delay time. The node synchronization method described.
5. The node synchronization method according to claim 4, wherein, when the slave node joins the round trip transmission delay time exceeding the margin time, the joining is rejected.
When a new node is joined to a plurality of nodes connected to the communication path or when the master node is dropped from the communication path, the synchronization between the master node and the slave node is re-executed. The node synchronization method according to claim 1, wherein:
The slave node measures the reception timing of the synchronization frame with the timer of its own node, and compares the measured time with a delay time at the time of synchronization with the master node set in advance. And a timer in the own node is synchronized with the timer of the master node so as to eliminate the above-mentioned synchronization deviation according to the detected deviation amount. The node synchronization method according to claim 1.
The node synchronization method according to claim 1, wherein the communication path is a star type having a relay device between the master node and the slave node.
A network transmission system in which a master node having a timer and a plurality of slave nodes are connected via a communication path, and the timers of the respective nodes are synchronized.
The slave node receives a synchronization frame transmitted by the master node, initializes a timer of the slave node, and transmits a reception completion frame to the master node.
The master node receives the reception completion frame transmitted by the slave node, and from the difference between the timer time at the time of reception and the timer time at the time of transmission of the synchronization frame transmitted from the own node, Calculate a round trip transmission delay time between the master node and the slave node, send a round trip transmission delay time notification frame to notify the slave node of the calculated round trip transmission delay time,
The slave node receives the round trip transmission delay time notification frame, and synchronizes a timer in the own node with a timer of the master node based on the round trip transmission delay time included in the notification frame. Network transmission system.
The master node is based on a maximum round-trip transmission delay time among the round-trip transmission delay times between a plurality of slave nodes.
The master node transmits the synchronization frame to all slave nodes, and sets a margin time from a timing at which all the slave nodes have transmitted a response to a timing at which the next synchronization frame is transmitted. The network transmission system according to claim 9.
The network transmission system according to claim 10, wherein the margin time is equal to or longer than the maximum round-trip transmission delay time.
The node newly requesting to join the communication path transmits a join request frame to the master node,
The master node, after determining whether or not to join the node requesting to join, when allowing participation as a slave node, transmits a join permission frame following the synchronization frame,
The node requesting to join receives the join permission frame and the synchronization frame, and as a slave node, transmits a reception delay frame to the master node at a reception timing of the next synchronization frame,
10. The master node according to claim 9, wherein a difference between a cycle timer value at the time of reception of the reception delay frame and a cycle timer value at the time of transmission of the synchronization frame is set as the round-trip transmission delay time. The network transmission system described.
13. The network transmission system according to claim 12, wherein if the slave node joins the round trip transmission delay time exceeding the margin time, the joining is rejected.
When a new node is joined to a plurality of nodes connected to the communication path or when the master node is dropped from the communication path, the synchronization between the master node and the slave node is re-executed. The network transmission system according to claim 9, wherein:
The slave node measures the reception timing of the synchronization frame with the timer of its own node, and compares the measured time with a delay time at the time of synchronization with the master node set in advance. And a timer in the own node is synchronized with the timer of the master node so as to eliminate the above-mentioned synchronization deviation according to the detected deviation amount. The network transmission system according to claim 1.
The network transmission system according to claim 9, wherein the communication path is a star type having a relay device between the master node and the slave node.
In a node device that synchronizes a timer of another node device connected to a communication path,
Network control means for setting either the master node or the slave node based on the connection state with the other node device;
When set to the slave node by the network control means, by receiving a synchronization frame transmitted by the master node, the timer of the slave node is initialized and a reception completion frame is transmitted to the master node System management means,
The system management means includes
When the master node is set by the network control means, the reception completion frame transmitted by the slave node is received, and the time of the timer at the time of reception and transmission of the synchronization frame transmitted from the own node are received. From the difference between the timer time of the hour, calculate the round trip transmission delay time between the master node and the slave node, send the round trip transmission delay time notification frame to notify the slave node the calculated round trip transmission delay time, A slave node receives the round trip transmission delay time notification frame, and synchronizes a timer in the own node with a timer of the master node based on the round trip transmission delay time included in the notification frame. Node device.
The system management means includes
Based on the maximum round-trip transmission delay time among the round-trip transmission delay times between a plurality of slave nodes when set to the master node by the network control means,
The master node transmits the synchronization frame to all slave nodes, and sets a margin time from a timing at which all the slave nodes have transmitted a response to a timing at which the next synchronization frame is transmitted. The node device according to claim 17.
The node device according to claim 18, wherein the margin time is equal to or longer than the maximum round-trip transmission delay time.
The system management means includes
When it is set as the master node by the network control means, it receives a join request frame from a node that has newly requested to join the communication path, and determines whether or not the node requesting to join can participate in the slave. When allowing participation as a node, a join permission frame is transmitted following the synchronization frame,
A cycle at the time of receiving the reception delay frame by causing the node requesting to join to receive the join permission frame and the synchronization frame, and as a slave node to transmit a reception delay frame at a reception timing of the next synchronization frame. The node device according to claim 17, wherein a difference between a timer value and a cycle timer value at the time of transmission of the synchronization frame is set as the round trip transmission delay time.
The network control means includes
21. The node apparatus according to claim 20, wherein when the slave node joins the round trip transmission delay time exceeding the margin time, the joining is rejected.
When a new node is joined to a plurality of nodes connected to the communication path or when the master node is dropped from the communication path, the synchronization between the master node and the slave node is re-executed. The node device according to claim 17, characterized in that:
The system management means includes
When set to the slave node by the network control means, the reception timing of the synchronization frame is measured by a timer of the own node, the measured time, and a delay time at the time of synchronization with the master node set in advance To detect whether a synchronization deviation with the master node has occurred, and in accordance with the detected amount of deviation, the timer in its own node is set so as to eliminate the synchronization deviation described above. The node device according to claim 17, wherein the node device is synchronized with a timer of the node.
The node device according to claim 17, wherein the communication path is a star type having a relay device between the master node and the slave node.