15
\$\begingroup\$

In a press statement, the Portuguese grid operator REN says, on the topic of today's Spanish-Portuguese large scale power outage (source)

It claims the outage that's affected Portugal was caused by a fault in the Spanish electricity grid, related to a "rare atmospheric phenomenon".

REN says that, due to extreme temperature variations in Spain, there were "anomalous oscillations" in very high-voltage lines.

It says this is known as "induced atmospheric variation", which in turn led to oscillations which caused synchronisation failures between systems.

That led to successive disturbances across the interconnected European network, it adds.

What is "induced atmospheric variation", and how would that lead to oscillations (I guess of power flow, but this is from the non-expert press, so who knows what they mean)?

EDIT: The oscillations are of "tension", i.e., voltage, as far as me guessing Portuguese works out; source: website of REN. REN says they occured in the Spanish side of the net, which lead to automatic net split.

O apagão que hoje assolou todo o território de Portugal continental decorre de uma significativa oscilação de tensões na rede espanhola num momento em que Portugal se encontrava a importar energia de Espanha. Com esta oscilação, os sistema de controlo e proteção das centrais portuguesas, e de acordo com o esperado numa situação com esta configuração, desligaram, dando origem ao apagão.

Translation:

The blackout that haunted the entirety of Continental Portugal's territory today was caused by a major voltage oscillation in Spain's electricity grid at a moment when Portugal was importing electricity from Spain. Due to the aforementioned voltage oscillation, the control and protection systems in Portugal's electricity grid were triggered, as expected of them during such anomalous circumstances, thereby creating the blackout.

\$\endgroup\$
11
  • 1
    \$\begingroup\$ I read that as an oscillation of the grid voltage or AC frequency induced by unusual weather-related demand and generation patterns. That is, the weather pushed energy flows in some unusual way that the grid could not handle while remaining synchronous. \$\endgroup\$ Commented Apr 28 at 20:30
  • 1
    \$\begingroup\$ Duplicate? physics.stackexchange.com/q/848666/299634 \$\endgroup\$ Commented Apr 29 at 7:39
  • 2
    \$\begingroup\$ I'm leaning on this term being the product of translation and communication failure. Blackouts are always cascading effect of some local disruption. Even rather sudden weather phenomena like Tornados don't form in seconds.. However, the damage and single point failures due to weather phenomena can happen as fast. So possibly it's just "wind" or "overtemp/overload shutdown" of a single transmission line triggering the cascading shutdowns. \$\endgroup\$ Commented Apr 29 at 7:43
  • 3
    \$\begingroup\$ @bobflux Not a great article though. "Grid instability suspected in unprecedented blackout" is as brilliant a headline as "virus suspected in Covid-19 pandemic". \$\endgroup\$ Commented Apr 29 at 10:36
  • 3
    \$\begingroup\$ As a native (brazilian) portuguese speaker, the original text from the REN website doesnt have much info. It just says that some significant voltage oscillations occurred in the Spanish grid while Portugal was importing energy, and as a result the grid was disconnected and they restarted the supply using a gas based generation plant. It doesnt even say if the event was generated by an atmospheric event or anything else, it just points at Spain. \$\endgroup\$ Commented Apr 29 at 16:09

5 Answers 5

Reset to default
11
\$\begingroup\$

It has nothing to do with atmospheric oscillations and this claim has since been withdrawn by authorities. It likely originated in a translation and/or comprehension glitch.

What actually seems to have happened is extremely similar to a situation which already occured in 2016 as analyzed by entsoe previously.

Then, just like now, there was:

  • lots of export from Iberia to France
  • ~0.2 Hz under-damped oscillation of the mains frequency in anti-phase to the rest of the continent.
  • Apparently, the frequency control encounters marginal stability. Traditional grids had a lot of implicit inertial response from heavy synchronous generators, which provide implicit local frequency coontrol. While this can be easily replaced by synthetic inertial frequency response from wind turbine VFDs or from battery array inverters in the future, we are in a transitional phase where a lot of gas fired steam turbines are taken off when there is renewable excess, but the frequency control reserve substitutes are still a bit weak. So I hypothesize that some TSO are cutting margins.

The difference between 2016 and now was, that the oscillation had a large enough amplitude this time, to get some part of the Iberian grid to go into automatic de-synchronization shutdown. This made the frequency control issue worse, spiralling out of control and leading to cascading shutdowns.

This cascade was contained near the Iberian-French interconnects, with the rest of Europe experiencing an intermittent frequency droop down to 49.8 Hz, but no shutdowns.

\$\endgroup\$
15
  • \$\begingroup\$ +1. As predicted in Redeia's annual report. Can be found by searching "Power outages due to high penetration of renewable energies" in this document: redeia.com/sites/default/files/publication/2025/02/downloadable/… \$\endgroup\$ Commented Apr 30 at 9:03
  • \$\begingroup\$ @bobflux Just want to add (not to shed bad light on renewables, which are great): It's an issue of the transitional phase in particular, and with PV in particular. When in conjunction with VFD wind turbines and batteries, PV is great. And synthetic inertia from batteries due to their large amount of energy (compared to traditional flyback energy) and peak power capabilities will end up being superior to traditional inertial frequency response. So while the future is bright, the dominant focus on (cheap) PV and negligence of wind and batteries is adding up to give rise to transitional issues. \$\endgroup\$ Commented Apr 30 at 9:31
  • 3
    \$\begingroup\$ I mean, that's the thing: the problem here was not that there's too much demand, or too little supply, or the supply cannot be controlled, or that there was sudden load or supply changes that a sensible control system couldn't have accounted for. The whole problem is just that the mechanisms of communicating load within the system are not adequate. So, that's 100% not a problem with cheap PV, but with technical debt. \$\endgroup\$ Commented Apr 30 at 13:53
  • 2
    \$\begingroup\$ @MarcusMüller you are right that we could switch to having a global time and grid phase setpoint and all nodes should just follow that, but this has issues with cold starts. For the time being with a centralized frequency control, adding more things that can source and sink lots of power for fractions of seconds provides damping for oscillation. PV is bad at this unless when not run at the maximum power point. So overall it's better to, run them at max power and provide the power flexibility via battery facilities or flywheel action of wind turbines, which are anyway necessary. \$\endgroup\$ Commented Apr 30 at 14:48
  • 1
    \$\begingroup\$ @MarcusMüller also, as long as there are steam turbines in the grid - and at least the nuclear fired ones will remain for many decades - you cannot change the grid frequency too much for their synchronous generators to work well. Again one could add VFDs, but not on the short/mid term. Plus there are many loads with synchronous motors. \$\endgroup\$ Commented Apr 30 at 14:51
4
\$\begingroup\$

What is "induced atmospheric variation"

An abnormal state (variation) of the operating parameters, induced by weather conditions. Specifically extreme temperature variations in Spain

and how would that lead to oscillations

The network, and loads, are both capacitive and inductive. If the generators and interconnects feed energy into the network at the correct phase and frequency, all is stable. If the network and loads return energy at the wrong phase, the system works like a giant spring, with energy passing back and forth between the loads, networks and generators. Oscillations are an expected characteristic of the system, in fact, the system is more efficient when run close to instability.

Voltage, power, current, and frequency are stabilized by feedback systems which respond to load phase and location, line capacitance, electronic control, rotating mass, and other factors. Like any control system, if it's run outside of it's expected input range, it may become unstable.

\$\endgroup\$
4
\$\begingroup\$

According to this article in Technology Magazine, at the time of the outage, only 28% of the grid was being supplied by non-intermittent sources such as nuclear, gas and hydro. The rest was coming from intermittent sources (60% solar and 11% wind).

"Atmospheric variations" such as clouds and lack of wind could result in rapid changes in energy flows which could destabilize a grid, triggering a latent failure mode. Major power outages in North America such as the Northeast blackout of 1965 and the Northeast blackout of 2003 have resulted from cascading failures initiated by a local issue.

Stability of complex power grids is a non-trivial subject.

\$\endgroup\$
3
\$\begingroup\$

Extraordinary interregional power flow, e.g. due to temperature differences or local failures. See ref. https://journals.aps.org/prresearch/abstract/10.1103/PhysRevResearch.7.013137, from Februar 2025. Spain appears to be particularly vulnerable. "As expected from our previous discussion, we can observe that the Pyrenees and the Balkans are the two most vulnerable areas of the system."

\$\endgroup\$
3
  • 2
    \$\begingroup\$ Welcome to EE.SE. Can you edit your answer to address the actual question, 'What is "induced atmospheric variation", and how would that lead to oscillations'? \$\endgroup\$ Commented Apr 28 at 20:03
  • 1
    \$\begingroup\$ Hi Fdhk645, how are extraordinary interregional power flow and the term "induced atmospheric variations" related? The paper you link to is interesting, but a generic simulation of the power grid and doesn't even mention atmospheric variations of any kind. (It starts with "due to man-made global warming of the atmosphere, we need to transform the power grid", but that's it.) \$\endgroup\$ Commented Apr 28 at 20:35
  • \$\begingroup\$ Thank you for the link. It's important to know where the weaknesses are so that the rarer effects can be considered for what would otherwise be 'minor' tripping issues. Icing, steady wind driven oscillation (see Tacoma Narrows..), and atmospheric swirls (cf dust devils) are all closely related rare but potentially catastrophic conditions. \$\endgroup\$ Commented Apr 29 at 8:09
2
\$\begingroup\$

Grid disruption caused by solar flares is typically a result of low frequency (effectively DC) voltages being induced in long power cables; this has the effect of saturating transformer cores and so the windings essentially become pieces of wire and short the supply. It’s easy to imagine that atmospheric conditions where thunderstorms are likely could create the same effect, particularly where there is a big change in altitude. I’d think that this is what is being referred to. Apologies, my Portuguese is limited at best.

\$\endgroup\$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.