IIDM 1.13: support of condensers

[annetill]

Describe the current behavior

For powsybl-open-loadflow and powsybl-open-reac, we need a way to make the difference between a generator that is not started and a generator that works as synchronous condenser. It is not possible for the moment, except forcing Pmin and Pmax of the condenser to be equal to zero.

Describe the expected behavior

With a new version of xiidm (1.13, 2024.2.0), we propose to add a field operatingAs or any proposal with following enum values:

• GENERATOR ;
• SYNCHRONOUS_CONDENSER ;
• MOTOR.

STATCOM not in that PR because it is a technology that we may want to model entirely.

Describe the motivation

No response

Extra Information

• Note that we all agree to say that this field is not dynamic and does not depend on targetP value of a generator for example.
• Note that an equipment can be GENERATOR and MOTOR depending on the exploitation time: this is the example of pumping stations.
• A generator could be GENERATOR and SYNCHRONOUS_CONDENSER in theory.
• A generator could be only GENERATOR, meaning that a strictly negative targetP is forbidden?
• A generator could be only MOTOR, meaning that a strictly positive targetP is forbidden?
• A generator could be only SYNCHRONOUS_CONDENSER, meaning that targetP is always zero.
• Could we imagine a generator MOTOR and SYNCHRONOUS_CONDENSER or MOTOR and GENERATOR and SYNCHRONOUS_CONDENSER?

[jeandemanged]

note: CGMES SSH equivalent is cim:SynchronousMachineOperatingMode enum with constants:

• generator
• condenser
• motor

[zamarrenolm]

I would choose MOTOR instead of PUMP for describing the generator is consuming active power (a pump is a specific type of consumption).

Do we need to make explicit in the base model the differences between synchronous condensers and STATCOM? From a high-level point of view, both equipment allow voltage regulation injecting/absorbing reactive power. If the difference is only relevant for dynamic analysis, both could be labelled as CONDENSER in IIDM. For steady-state analysis they will behave similarly, but could have different dynamic models.

[annetill]

Update: finally, after a talk with @jeandemanged, it seems that a boolean synchronousCondenser to true or false is enough to treat the case that a generator is controlling voltage even if its targetP is equal to zero.

[annetill]

Update: finally, after a talk with @jeandemanged, it seems that a boolean synchronousCondenser to true or false is enough to treat the case that a generator is controlling voltage even if its targetP is equal to zero.

As I think users will use that field to model STATCOM too (https://www.hitachienergy.com/us/en/products-and-solutions/facts/statcom), even if a good modelling could be to rework on StaticVarCompensator later, I think we have to remove "synchronous" from boolean name. Maybe juste "condenser" or "compensator"?