JPH09319407A - Time synchronizing method for distributed system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a time synchronizing method for distributed system with which the time can be synchronized with high accuracy between slave controllers. SOLUTION: Concerning this distributed system, time information expressing a time longer than minutes is periodically transmitted from a main controller (host controller) 1 to input/output cards (slave controllers) 2a-2d, the time longer than minutes is synchronized among the respective input/output cards 2a-2d, a prescribed synchronizing signal is periodically transmitted from the input/ output card 2a to the other input/output cards 2b-2d, and the time shorter than seconds is synchronized among the respective input/output cards 2a-2d.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a time synchronization method for a distributed system composed of an upper controller and two or more lower controllers.

[0002]

2. Description of the Related Art In a distributed system composed of an upper controller and two or more lower controllers, it is necessary to synchronize the time among the lower controllers. For this reason,
Generally, a method of transmitting time information from a higher-order controller to a lower-order controller and synchronizing the time between the respective lower-order controllers is adopted.

Specifically, when the conventional distributed system is applied to a plant control device such as a nuclear power plant or a thermal power plant as shown in FIG. 2, a main controller 11 as an upper controller to a plurality of input / outputs as lower controllers. Time information is periodically transmitted to the card 12, and the time is synchronized between the input / output cards 12 based on this time information.

However, the time information transmitted from the main controller 11 to each input / output card 12 has a certain transmission delay, and this transmission delay differs depending on the transmission distance from the main controller 11 to the input / output card 12. It is difficult to synchronize the time between the input / output cards 12 with an accuracy equal to or less than the transmission delay, for example, an accuracy equal to or less than milliseconds.

[0005]

As described above, in the conventional distributed system, since the time information is transmitted from the upper controller to the lower controller to synchronize the time between the lower controllers, there is a delay in transmitting the time information. However, there is a problem that it is difficult to synchronize the time with an accuracy of millisecond or less. It is an object of the present invention to provide a time synchronization method for a distributed system, which can perform time synchronization between lower controllers with high accuracy.

[0006]

In order to solve the above-mentioned problems, the present invention provides a time synchronization method for a distributed system comprising an upper controller and two or more lower controllers, in which the upper controller has two or more lower controllers. , The time information indicating the time of more than a minute is periodically transmitted to synchronize the time of more than a minute between the lower controllers, and one lower controller preset from two or more lower controllers to another lower controller. Is characterized in that a predetermined synchronization signal is periodically transmitted to synchronize the time less than a second between the subordinate controllers.

In the time synchronization method of the distributed system having such a configuration, the time of minutes or more between the lower controllers is synchronized by the time information transmitted from the upper controller to the lower controller, and the time of seconds or less is controlled between the lower controllers. It is synchronized by the synchronizing signal transmitted. Therefore,
Even if there is a transmission delay in the time information transmitted from the upper controller to the lower controller, the transmission delay is 1
If it is within a minute, time synchronization of more than a minute can be achieved between the respective lower controllers, and time synchronization of a second or less can be achieved by a synchronization signal transmitted between the controllers separately from this processing.

As described above, by dividing the processing for synchronizing the time between the respective lower controllers into minutes or more and seconds or less, the time synchronization between the lower controllers can be performed with high accuracy.

[0009]

1 shows a system configuration when a time synchronization method for a distributed system according to an embodiment of the present invention is applied to a plant control device such as a nuclear power plant or a thermal power plant. , A main controller 1 which is an upper controller and input / output cards 2a to 2d which are lower controllers. Here, each of the input / output cards 2a to 2d is capable of managing time less than a second with a resolution of 1 millisecond by a timer using a crystal oscillator. Further, the input / output card 2a is a reference controller for synchronizing the time between the input / output cards 2a to 2d in seconds or less, and is provided with a pulse output circuit 3 for outputting a predetermined contact pulse. Further, the other input / output cards 2b to 2d are provided with a pulse input circuit 4 to which this contact pulse is input.

The main controller 1 is composed of input / output cards 2a-2
Time information indicating a time longer than a minute is transmitted to d every minute.
The input / output cards 2a to 2d synchronize the time of minutes or more based on this time information.

On the other hand, the input / output card 2a, which is the reference controller, outputs a contact pulse, which is a synchronizing signal, to the other input / output cards 2b to 2d via the pulse output circuit 3 every 60 seconds. The input / output cards 2b to 2d include the pulse input circuit 4
The time less than a second is synchronized with the contact pulse input from. By doing this, the input / output cards 2a-
It is possible to achieve time synchronization of 2 seconds or less between 2d.

That is, in this embodiment, the input / output card 2
The processing for time synchronization between a to 2d is divided into minutes and more and seconds and less, minutes and more times are synchronized by time information transmitted from the main controller 1, and times less than seconds are input / output cards 2a to It is synchronized by the contact pulse transmitted between 2d.

Therefore, even if a transmission delay occurs in the time information transmitted from the main controller 1 to the input / output cards 2a to 2d, if the transmission delay is within the transmission cycle of the time information, that is, within 1 minute, Time synchronization of minutes or more can be achieved, and time synchronization of seconds or less is performed separately from this processing. That is, the entire system is not affected by the transmission delay of the time information.

Further, as described above, the input / output card 2a
Since ~ 2d manages time with 1 millisecond resolution,
These time synchronizations can be achieved with millisecond precision, and, for example, processing that requires time resolution of 1 millisecond can be distributed to the input / output cards 2a to 2d.

As described above, the time synchronization method of the distributed system according to the present embodiment has higher accuracy than the conventional method of synchronizing the time of the lower controller based only on the time information communicated from the upper controller. Since it is possible to synchronize time with, it is effective when it is applied to applications that require particularly high accuracy for time synchronization of distributed systems, such as plant control of nuclear power, thermal power or chemicals, or event recorders. .

[0016]

As described above, according to the present invention, the time synchronization between the lower controllers can be performed with high accuracy by dividing the processing for synchronizing the time between the respective lower controllers into minutes or more and seconds or less. It is possible to provide a time synchronization method for a distributed system that can achieve the above.

[Brief description of drawings]

FIG. 1 is a system configuration diagram for explaining a time synchronization method for a distributed system according to an embodiment of the present invention.

FIG. 2 is a system configuration diagram for explaining a conventional time synchronization method for a distributed system.

[Explanation of symbols]

 1 ... Main controller 2a-2d ... I / O card 3 ... Pulse output circuit 4 ... Pulse input circuit 11 ... Main controller 12 ... I / O card

Claims (1)

[Claims] 1. A time synchronization method for a distributed system comprising an upper controller and two or more lower controllers, wherein the upper controller periodically sends time information indicating minutes or more to the two or more lower controllers. To synchronize the time of more than a minute between each lower controller, and periodically transmit a predetermined synchronization signal from one preset lower controller of the two or more lower controllers to another lower controller. And a time synchronization method for a distributed system, characterized in that each subordinate controller synchronizes time less than a second.