There is no mechanism proposed by the Standard to positively identify data points in the data model. For instance PTOC can refer
SCL of IEC 61850-6, Ed. 2.1 will support the possibility to relate logical node instances to function / subfunctions / eqfunction / eqsubfunction. There is a new task force in WG10 that will specifiy Function / Subfunction types. Once Function / Subfunction types are defined and used in SCL, this will allow to link a LN Instance to the particular Subfunction.
It is recommanded in addition to use an approriate description attribute of the LN data model.
aAddressed in the 61850-6-100
6-100 and Osmose are not enforcing directly additional semantics to the IED data model, but the definition of the ISD file together with the data exchange and Power System Resource Reference allow the user to specify semantics in a much more formalized way:
1. SourceRef and ControlRef allow to add semantic information by:
a. Providing naming fields for the user to add some meaning
b. Create references to co-operating LNodes, e.g. CSWI and XCBR can be linked via SourceRef
2. Power System Resource References create a link between a Function/SubFunction and any element of the Substation/Process section. With this the user can express e.g.
a. Association of MMXU and CT or VT
b. Circuit Breaker which is tripped by a PTRC
c. Line which is protected by a PDIS
3. Applications are introduces in the Substation/Process section. They relate Functions and in this way also Logical Nodes to a common semantic. E.g. if an RBRF and a PTOC belong to the same Application, it is obvious that it provides the initiation of the RBRF
All these semantic refinements take place in the Substation/Process section. 6-100 also gives guidelines how to create an ISD file out of this specification and allows in this way to transfer this semantic to the relay selection/procurement.
6-100 has improved the modeling, but has not yet standardized the Function sand Applications.
(1) Standardising function / subfunction types as mentioned in Issue 4 will help to support the semantic identification of vendor instantiated LNs
(2) Modeling guidelines / agreements may further help
There is no mechanism proposed by the Standard to positively identify data points in the data model. For instance PTOC can refer to any number of protection events related to current. Does User have to ask Vendor to make indexing and prefix arrangement as he (User) would specify for each generic LN that covers functionality segregated to several instances(for example definite time OC Dt1EftPTOC1, Dt1PhsPTOC2,.) ? And how will software that will import/communicate model recognize which DO has semantics that is required? Should description be mandatory part of import process during configuration of Client and how it should be integrated (rules)? More precise LN modeling rules required.