Who Done It: The Spanish Blackout Mystery
Like Murder Mysteries, Answers To The Blackout Only Start When The Lights Come Back On
Every good detective knows they’ll never solve a murder mystery on the first day, or even the second. The “who done it” only reveals itself after days of careful investigation and drawn-out deliberation. Power outages, it turns out, are no different.
Just after 12:30pm on April 28th, Spain and Portugal suffered a massive blackout that left most of their population without power for over ten hours. This is a big deal, but luckily such events are exceedingly rare, and energy restoration happened relatively smoothly. In fact, it’s been over 20 years since an event of such magnitude happened in the U.S., when a huge blackout left most of the Northeast powerless in August 2003.
There’s still much we don’t know about why the Iberian Peninsula went dark, but we can be sure of one thing—anyone claiming definitive answers or assigning blame at this time doesn’t have all the details and is rushing the timeline. For example, an official account of what happened during the 2003 Northeast blackout wasn’t published until eight months after the event and it ran 238 pages.
We likely won’t know what happened in Spain for a good while. The event happened quickly, which makes investigating the cause much harder for two reasons. First, investigators will need to compare timestamps from various equipment logs with clocks that are necessarily well synchronized. This means it will not always be easy to know the sequence of events, making it more difficult to develop a solid chain of events. Second, when parameters like voltage, frequency, and current change quickly on the power grid, the corresponding power-flow math becomes much more complicated (harder to make simplifying assumptions).
Another thing to keep in mind with major blackouts is that there is usually a proximate cause. In the U.S. Northeast case, sagging transmission lines sparking with tree branches and a software bug in the alarm system at the control room of FirstEnergy, combined with a whole context of poor planning and operating practices, led to subsequent cascading failures. It usually takes even longer to sort out chains of events like this.
All this to say, anyone wanting to solve this “who done it” with what we know right now is as likely to solve the case as someone who has only read the first ten pages of a mystery novel. Making definitive claims at this time is likely to leave you on the wrong side of a plot twist.
What we know
Some widely reported facts about what happened can at least help us begin to piece together what caused the lights to go out. Here’s what we know so far, although our understanding of this event is sure to evolve as more information emerges.
WHO: Spain’s electricity grid became unstable, causing power plants to trip offline. Which specific plants tripped offline first, and what caused this instability in the first place, remain unknown. In the minutes leading up to the outage, there was plenty of electricity supply to go around. This was not a problem of “the sun didn't shine, or the wind didn't blow:" Spain was exporting gigawatts (GWs) of power to France, Portugal and Morocco, as well as 3 GWs to pumping loads (most likely water to one of the 6 GW of pumped hydro storage reservoirs).
WHAT: Nobody yet knows what caused Spain’s grid to become unstable, and why that instability cascaded into a widespread blackout. At 12:33pm on April 28th, something started a cascade of power plants tripping offline, and in under five seconds Spain had lost over 15 GW of generation (just over half of what was online). Grid frequency oscillations had reported on the European grid in the half-hour leading up to the event, so these could end up an important part of the story that emerges as well. But the cause of these oscillations, and their role in the blackout, is part of the mystery that remains to be solved.
WHEN: The critical time period appears to be from around noon – when grid frequency oscillations first appeared – to 12:33, when large amounts of generation tripped in a matter of seconds. Then came the long and complicated process of restoring power, re-energizing parts of the grid that had completely lost power, and slowing bring generators and demand back online. Nearly 24 hours passed before Spain’s grid was back to normal operating conditions.
WHY: Why did this happen, and what should we do to prevent a similar event from happening in the future? This is where we simply do not yet know, and likely will not for some time. Without a clear understanding and explanation of this event, many commenters have filled the void with a variety of theories. Only time will tell which ones are supported by the evidence.
For instance, some have argued that solar started the cascade of power plants tripping offline, potentially because of sensitive power electronics. Since solar farms were 53 percent of Spain’s generation at the time of the event, they are an easy target to point the finger at. But each individual solar farm is small relative to the size of the grid (most solar farms in Spain, roughly 80 percent of the solar capacity, are sized at less than 50 megawatts), and so far, there is no explanation for a sudden simultaneous collective failure of a large amount of solar capacity.
Some have pointed to a lack of inertia – the ability of large spinning generators to slow down a drop in the grid’s frequency – as a possible cause. With only 30 percent of the system generation coming from synchronized generators with a spinning mass, it’s possible that more inertia or fast frequency response from battery energy storage could have slowed down the cascade of plants tripping offline, but it remains to be seen if this was a factor at play in this event.
What next?
We need to let investigators do their work, so we can find the causes (both proximate and wider context) and put in place solutions to make sure this doesn’t happen. This is certainly no smoking gun against a renewable heavy system, even if inertia ends up being the ultimate culprit. Modern power systems can certainly go well below 30 percent synchronized generation if they have fast frequency response resources to support the grid (and the monetary incentives or ancillary markets to pay for them).
Fast frequency response from storage is already contributing in regions like California, Texas, Ireland and South Australia with high penetrations of inverter-based resources. Spain was already targeting the installation of these resources by 2030. As further synchronized generation drops off the grid, new solutions like batteries with grid-forming inverters can provide even deeper support.
Whatever we do, grid operations will change because of new technology, evolving load patterns, changing climate, deteriorating legacy systems, and workforce turnover. Exceptional events and human error will reveal our vulnerabilities and cause outages and brownouts. It’s important to react soberly to such events, find out what happened and develop fixes. Expert groups like the Energy Systems Integration Group also work tirelessly to collect data and expert know-how to show policy makers how to preempt possible problems. For more technical details and expert comments on the blackout, see Carbon Brief’s excellent Q&A.
 | A guest post by
|
 | A guest post by
|