1257   Partial ExtRefs

Created: 08 May 2014

Status: In Force (green)

Part: Part 6 (2009-12; Edition 2)

Links:

Page:

Clause: 9.3.13

Paragraph:

Category: Issue for edition 3 of this part

Issue

Part 6, Clause 9.3.13, explicitly allows only two forms of ExtRef elements:
1) Full (resolvable) reference to the input signal(s), (i.e., iedName, ldInst, prefix if there is one, lnClass, lnInst, doName and daName, which can be omitted in case of an FCD), optionally intAddr, desc, serviceType and reference to the sending control block.
2) Just the intAddr and optionally desc and serviceType, typically in the ICD file case. The desc would contain human-readable information that will help the IED engineer "connect" the input signals.

I propose to add a 3rd case, also for ICD/IID files (or partially engineered SCD), where one could add more semantic information about the input signal(s): to have just LN class, DO name, DA name, intAddr, and optionally desc and serviceType

The reference to the signal would be only partial (no or optional iedName, ldInst, prefix, and lnInst) and describe only the semantics. This way, tools can help do the mapping of input signals to internal signals and help detect inconsistencies/errors.

Of course, during engineering this information might be replaced (e.g., use a XCBR.Pos.stVal instead of a CSWI.Pos.stVal).

Proposal

Describe this additional case in Part 6.

Discussion Created Status
15 Oct 14 In Force (green)
Overall needs and usage in the engineering process will be taken up for Edition 3. No change at the moment for Edition 2. 11 Sep 14 Ballot Period
However, an ICD from a device may not know that it needs what type of LNClass that is needed. It may only know it needs a POS of a specific type of CDC.

As another example it may know that it needs PHV.Phs.instmag (MX). Therefore, I beleive there needs to be a mechanism to specify anything level (upward or downward) of a LNClass. It also needs to capability of specifying an FC and CDC/datatype expected.
04 Jun 14 Discussion (red)
A prefiltering of what should/could/must/shall be used as external reference is an interesting issue. We must however be careful of the impact of such input "preparation" because of the BlkRef, InRef that also provide semantic information of the destination. The content of InRef7BlkRef are object refernece and should not include only lnClass / do / da.
We should should further discuss the system aspects for enhancing the system engineering and standardize it in Ed3.
09 May 14 Discussion (red)

 

Privacy | Contact | Disclaimer

Tissue DB v. 23.12.13.1