Often people ask the question: Where should I allocate a specific function (and the corresponding Logical Node) in the hierarchy of: process, bay, substation, regional control center, central control center?
In the following you will find some description from the Standard IEC 61850-5, a paper from 2001, a question I received the other day, my own brief answer to that question, and two answers from two good friends: Andrea Bonetti (Megger) and Joachim Lange (Solvay).
IEC 61850-5 (Communication requirements for functions and device models; Ed 2022) describes that the allocation is free to allow different architectures and levels were a function (respective a corresponding LN) can be allocated (means implemented).
Excerpt of Clause 9.2.1 Free allocation of Logical Nodes
"The free (arbitrary) allocation of functions or Logical Nodes respectively is not restricted to the common level structure."
The following excerpt of clause 10.1 Need for a formal system description is one of the crucial clauses in the whole standard series IEC 61850:
"Where the data is coming from (sending Logical Node) and is going to (receiving Logical Node), i.e. the static structure of the communication system, has to be engineered or negotiated during the set-up phase of the system. All functions in the IEDs have to know what data to send when and what data they need from functions in other IEDs to be able to fulfill their functions. To control the free allocation of functions respectively Logical Nodes and to create interoperable systems, a strong formal device and system description for communication engineering shall be provided. Such a description (System Configuration description Language) is defined in Part 6 of this standard (IEC 61850-6). This formal description shall also support the data exchange between different tools if applicable."
Be aware that the Logical Nodes are to be understood as a wrapper around a function. In most cases the function as such is behind the facade of the Logical Node. One exception is the Logical Node class FSCH (Schedule). The definition of FSCH contains a well defined state machine that is part of the function of a scheduler.
An old paper from the year 2001 may help you to understand the approach of IEC 61850
The Impact of the coming Standard IEC61850 on the Life-cycle of Open Communication Systems in Substations
By Lars Andersson, Klaus-Peter Brand, Wolfgang Wimmer; ABB Power Automation Ltd., Switzerland
Excerpt from the paper:
- Free allocation of functions [KHS: and therefore free allocation of Logical Nodes]
- Extension rules to support new functionality
- Separation of communication from application issues in a well defined manner
- Description of the station from the application communication point of view.
Click HERE for the paper published in the year 2001.
Question:
Hi Karlheinz,
I’m a system engineer with a question on IEC 61850 in substation automation.
In a ring with all IEDs and two RTUs, are the RTUs only SCADA gateways, or can they also host SAS control logic (e.g., with a T500’s basic logic capability)?
Should a SAS operate autonomously from SCADA or higher-level PLCs, and if so, should the IEC 61850 RTUs implement control logic for outage restoration, load shedding, etc., to ensure autonomy?
Answer from Karlheinz Schwarz
Dear xx,
Thanks for contacting me.
IEC 61850 is independent from centralized or decentralized approach. It depends on the philosophy of the utility how to architect the system. Functions could be in the multi-functional IED (Relay), bay controller, substation controller, SCADA, control center, ...
IEC 61850 may be used to run schedules in control IED right behind the electrical connecting point of a home, factory, ... using the LN FSCH - Scheduling.
In Germany we have the so-called FNN Steuerbox that uses schedules for limiting the power usage ...
Hope that helps.
Answer from Andrea Bonetti
IEC 61850 does not prescribe where control logic must be located (read it as the famous sentence "free allocation of the Logical Nodes").
The decision is up to the system designer and the utility’s operational requirements. If autonomy of the HV ring is desired, logic may be placed in RTUs, bay controllers, or other IEDs so that the system operates without SCADA. Any such requirement would come from utility or regulatory specifications, not from the IEC 61850 standard.
There is no IEC 61850 requirement that mandates where control logic must be located — whether in RTUs, IEDs, or higher-level systems.
IEC 61850 specifies how devices exchange information and how to engineer all of that (SCL engineering), not where the logic resides.
Whether the HV ring is autonomous is purely a system design choice defined by the utility’s operational philosophy, national regulations, or internal standards — not by IEC 61850 itself.
If autonomy is required (e.g., for outage restoration or load shedding without SCADA), the designer can choose to implement logic in RTUs, bay controllers, or other IEDs so they can function without higher-level supervision.
Obviously it depends also on the voltage level. Usually, higher voltage level –> less integration. Lower voltage level à more integration.
But there are exceptions to this rule like always.
Answer from Joachim Lange
In case of classical terminals, neither the terminal number nor the terminal function is defined in any standard. Personally, in case of CFC implication I use UDx baycontrol blocks, defining „my signals“ and a group with the "GOOSE exchange" signals
- in case of blocking signals like in double busbar structures I do this as well, because I use busbar selective reverse blocking in dependance of position information,
- this means that disconnector positions enables/disables the blocking transmission to its circuit breaker protection.
It is even useful to avoid that a send out blocking signals triggers during test a not involved feeder.
Some grid companies use f.e. blocking signal in combination with breaker failure. This means when signal is not reset in time they trigger the breaker failure protection.
So the CFC function is really case wise.
We have in our house in the UD1 group all signals which are used for bay supervision.
Philosophy: I decentralise load shedding and automation functions into the bay control level ( discrete frequency / voltage levels with hysteresis).
Such bay controller measures autonomously its conditions.
I provide from above (Scada) the enabling/ disabling signals or mode selections ( power level ) or setpoint correction signals.
The advantage is that a single device failure may not impact a hole system.
Note that Andrea Bonetti (Megger), Joachim Lange (Solvay), Dr. Ghada Elbez (KIT), and I will conduct a comprehensive training starting 09.-13. March 2026 Karlsruhe (Germany) and 21.-25. September 2026 Karlsruhe (Germany).
We will provide the details in the next weeks. Stay tuned.
Here are the logos for that training:
Posts
