323 Goose Retransmission Issue
Created: 16 May 2006
Status: In Force (green)
Part: Part 8-1 (2004)
Issue: Is the publisher supposed to re-transmit all Goose Messages sent previously or whether it should re transmit only the last message ?
If it is to retransmit only the last message ; and if there were to occur some very fast changes of data in the dataset(lets say Binary data toggling back),it will trigger Goose report to sent at very short intervals of time with stNum incremented for each value change. In that case,there is a reasonable probablity that some of the messages may be lost since the goose messages are multicast messages for which there is
no guarantee of delivery.
Proposal: In my opinion since IEC 61850 is supposed to be complete standard for the entire substation, the standard must comment on these issues.
Note: To see attachments you have to log-in first.
Added slight modification to proposed state machine.
||02 Jun 06
It is a little unclear in the state diagram in Figure 9 (figure is mislabeled and has been the subject of another tissue).
I propose to change the state diagram to reflect what is posted in the attachement. Where state (5) is the local detection of a value change.
This should clear up the confusion.
|24 May 06
The GOOSE communication service is intended to distribute status information accross IEDs. The subscribers of GOOSE messages are interested in getting as quick as possible the latest status information. Since the communication services used are unconfirmed, there is no reason to repeat any older messages. GOOSE services are not adequate to create an event history based on the GOOSE events.
So the answer is: the publisher is only supposed to transmit the current status of the information in the GOOSE message. As such, it is not really a retransmission of a message, it is a cyclic transmission of status information, where the cycle time is not fixed, but changing depending on changes in the status information.
|16 May 06
Privacy | Contact | Disclaimer
Tissue DB v. 184.108.40.206