Reactive power exchange between eTraGo and eDisGo

[openego]

Reactive power exchange between eTraGo and eDisGo

[maltesc]

First I think, it is important to distinguish the reactive power management with regard to the different scenarios. For the status quo scenario I would suggest to focus on today's practices. For the NEP 2035 and especially the eGo100 scenario, the circumstances are likely to change.

Status Quo

Today, reactive power control (and thus voltage control) in the transmission grid is realized by synchronous generators and compensating inductors for long transmission lines. Furthermore, the transmission system operators (TSOs) not only provide reactive power for the transmission grid, but also for the distribution grids [1].
Today, the Q balance (delta Q) at each MV-HV transition point is not limited and it is fully covered by the reactive power management of the transmission grid. In times of high wind feed-in, already today this results in a high inductive power demand of distribution grids [1].
Today, the P/Q control of photovoltaics generators is only used for local voltage maintenance but not in order to support the local Q balance.

Should this delta Q be considered in eTraGo (with regard to PF post LOPF method)? I think we decided that this is not desired and will be disregarded. However, it would be fairly easy to implement...

NEP 2035 and eGo100

In the future, the situation is likely to change. With a decreasing number of synchronous generators and a simultaneous increase of DGs in distribution grids, TSOs will have to rely on reactive power from distribution grids. A good analysis of the future reactive power exchange between distribution and transmission grids can be found in [2]. Generally speaking, here it is assumed that TSOs send reactive power set points to DSOs with the goal to attain a steady voltage in the transmission grid. DSOs then 'try' to reach these set points under compliance with the operational constraints of distribution grids (in case of the study these are: max 1.08 pu voltage and 70% thermal capacity [2]). In the study, a distribution management tool is performing this optimization task. I think, a similar behavior would be desirable for eGo's MV-HV transition points.

However, the question arises of how this could be implemented in eGo. A discussed solution was to include the reactive power management in the eTraGo simulations and subsequently use the fixed P/Q values for every generator and storage unit as inputs in eDisGo. This assures a DG-based voltage maintenance in the transmission grid, however disregards the local operational constraints of distribution grids. This will probably not find an optimal solution.

In accordance with the study (cf. [2]), it would be more realistic to calculate 'desired' Q set points with eTraGo at every MV-HV transition point and then MV grids try to meet these set points within the operational constraints. However, I think this is complex and time-consuming to implement.

In order to model the reactive power exchange between MV and HV/EHV grids in open_eGo, this topic should be further discussed. This text is not comprehensive, but might serve as a fist orientation. Please feel free to add any suggestions and ideas...

Literature

[1] Kämpf, E., Schmidt, S., Walther, B., Wildenhues, S., Eggemeyer, R., Brantl, J., & Braun, M. (2013). Einhaltung definierter Blindleistungsbänder an HS/MS Übergabestellen durch Einsatz der Blindleistungsfähigkeit dezentraler Einspeiser. In Proc. Internationaler ETG-Kongress 2013: Energieversorgung auf dem Weg nach (Vol. 2050).

[2] Marten, F., Diwold, K., Lower, L., Faiella, L. M., Hochloff, P., Hansen, L. H., & Braun, M. (2013, November). Analysis of a reactive power exchange between distribution and transmission grids. In Intelligent Energy Systems (IWIES), 2013 IEEE International Workshop on (pp. 52-57). IEEE.

[boltbeard]

I think the current assumptions that have been harmonized between eTraGo and eDisGo are in the the open_eGo redmine under the Sub-workpackage #935 (in clear text ..../redmine/issues/935). We could double check this and see if something could be modified to help with the NEP2035 and the eGo_100 situations

[wolfbunke]

Should this part of the RTD docu ? @ulfmueller @maltesc