Renewables and Transient Stability

[element230]

Dear all,

I’m trying to do a simplified analysis on how the integration of renewables into the transmission system would impact stability and critical clearing times.

Consider this -- Most of renewables have limited (if any) inertia and have also limited ability to respond to system disturbances and contingencies. In terms of the loss of inertia, I assume that the integration of more and more renewables (in particular wind and PV), will create a more unstable system, as these types of technologies are”behind electronics” and thus do not have a direct coupling with the system.

How I’m thinking of doing this analysis? My utility works stability modeling with PSS/e. This is a great program, but given the current state of mathematical models available for wind and PV plants, the modeling of multiple (say, tens) of these facilities is cumbersome and complicated. What I was thinking of doing was that instead of modeling these facilities by their corresponding mathematical model, I represent them as inertia-less classical generators (as I’m interested in simulate the lack or decrease in system inertia). One other way that came into my mind was to model the as negative load (corresponding to their name plate capacity), but I think this will yield in some unintended effects on the simulation.

What are your thoughts about my approach? Do you have better ideas?

As always, thank you for your comments!!!!

 

[juleselec]

By transient stability, I assume that you are talking about rotor angle stability due to large disturbances (or first-swing stability). As you point out, renewables sitting behind power electronic interfaces will not contribute any inertia to the system, but as far as transient stability is concerned, this isn't necessarily a bad thing. Consider that with less overall inertia in the system, there is less kinetic energy developed by the rotating machines during a fault / contingency. Therefore, machine rotor angles will not increase as much. In terms of the equal-area criterion, it means that the areas under the power-rotor angle plot are smaller with lower system inertia.

However high penetration of renewables have adverse impacts on other forms of stability, e.g.
[ul]
[li]Short-term voltage stability, especially for older types of wind turbines without reactive power support during faults. Basically with higher penetration, there is lower short circuit power in the system and thus it is more prone to larger voltage sags.[/li]
[li]Long-term voltage stability, since renewables are typically connected at relatively weak points to the transmission (or sub-transmission) network. The re-alignment of power flows after a contingency (e.g. a loss of a conventional power plant) can lead to voltage stability problems throughout the system, i.e. parts of the system are near the nose points of their respective PV curves.[/li]
[li]Frequency stability, which is related to the lack of inertia of renewables and the fact that they are not used for primary frequency control. Large contingencies (e.g. loss of a main interconnector or power plant) can lead to frequency stability issues when there is higher penetration of renewables, such as large drops in frequency (depending on the size and strength of the network).[/li]
[/ul]

Anyway, regarding your modelling question, you should determine what kind of renewable sources that you have in the system and whether they contribute any reactive power support during a fault. If not, then modelling them as negative PQ loads is acceptable. In any case, you probably won't see much difference for a transient stability study.

 

[Adam1980]

Hello,

you are totally correct concerning the problem of contribution of non-rotating generation such as PV or decoupled generation which is actually all generation with an inverter interface to the grid to the inertia of the grid.

Some grid codes are starting to tackle this issue but most are related to active power ramp down when the frequency increase according to a pre specified dp/df. I would recommend that you investigate the condition in Ireland since there is much discussion about this topic there right now due to the fact that the system is islanded with a high penetration of wind.

Some wind generators manufactueres are developing "virtual inertia" for their products.

http://www.icrepq.com/icrepq'11/510-torres.pdf

(general literature).

However for modelling purposes I believe you should include the dynamic model of the wind turbine/PV as well as plant level controllers.

 

[element230]

Thank you for your comments. Both are very valuable and provide great points of view.

I think I'm going to do both and see what comes out of this. Our situation is similar to Ireland in the sense that we are an island, with relatively low inertia. Our renewable integration is expected to be significant (about 50% in nameplate capacity, 20% energy) in a very short time 2-3 yrs.

With regards to ancillary services support, we are requiring facilities to abide to minimum technical requirements that provide for things such as voltage/reactive control, frequency support (both up/down regulation) and voltage ride through.