Blackout lessons: Stability challenges for modern grids

Modern power systems and markets are more complicated than ever. Three priorities must be kept in balance: economics, politics and physics. Firstly, electricity must be affordable for consumers, and only the cheapest generation technologies remain competitive in the market. Secondly, governments face pressure to keep prices stable and work towards rapid decarbonisation. Finally –and most importantly – the system must always obey the hard rules of alternating current (AC) power physics.

Grid frequency matters 

Let’s get into the details (but don’t worry, I promise this is as technical as it gets). All national grids operate on AC. In Europe, for example, the power alternates or reverses direction 50 times a second (50Hz). Grid operators closely watch this frequency: if it falls, demand exceeds supply and if it rises, generation is too high. They rely on fast-acting power stations to quickly adjust their output and keep the grid stable. 

However, AC power physics is unforgiving. 

Throughout the mid-20th century, grids were dominated by large power stations burning coal, gas, oil or splitting uranium. If one unit failed, the inertia from the spinning turbines in these plants slowed any frequency drop, giving time for corrective action. But when inertia is low, the system is less able to react to sudden changes in supply or demand. 

Now, more power is generated through renewables. Solar and wind don’t provide natural inertia, except in certain rare cases. So unlike traditional spinning turbine generators, these systems don’t naturally support grid stability, as they aren’t directly synchronised with the grid’s frequency. As renewables grow, grid operators must find new sources of inertia: large spinning machines (like synchronous condensers) or advanced power electronics creating synthetic inertia.  

Renewables are often connected to the grid via power electronics, such as inverters, which convert the electricity to the standard AC supplied to the grid. Much of today’s renewable energy runs through “grid following” inverters, which require a steady grid frequency to function. Many are also connected at the local distribution level, which brings extra safety challenges. For example, if a cable gets damaged during construction works but a small local generator keeps running, it could stay live and dangerous. To mitigate this, inverters are programmed to detect rapid frequency changes and shut off. However, on grids with low inertia, sudden changes can trip many inverters at once, causing cascades of disconnections that risk wider blackouts. 

Frequency drops can also cause automated systems to disconnect substations to reduce demand and help stabilise the grid. But with lots of distributed generation, disconnecting substations may simultaneously remove crucial local generation, sometimes making matters worse. 

That’s why inertia really matters. 

Adding inertia 

Today, many grids use synchronous condensers – large, motor-driven machines that add inertia without producing electricity. Although not new, their role is becoming more critical. The other option is “grid-forming” inverters, synthetic solutions that mimic the stabilising effects of traditional machines. This technology is advancing, but it’s more expensive and less mature. 

Operators measure inertia in several ways, but a handy benchmark comes from Ireland: system non-synchronous penetration (SNSP). To calculate SNSP you take the proportion of non-synchronous generation as a proportion of total system demand. 

The SNSP risk scale is roughly: 

• Low risk: below 50% – lots of conventional generation 

• Medium risk: 50-75% – requires synthetic inertia and fast-acting reserves 

• High risk: above 75% – higher risk of instability and possible blackouts if not managed carefully 

Interconnections and island grids 

Interconnection between national power grids is also an integral part of this issue. There are two main ways countries connect their power grids. AC interconnection can share inertia across borders. But high-voltage direct current (HVDC) links – often running under the sea – cannot as they require big power converters at each end to convert HVDC into AC and vice versa. 

The greatest challenges are for “island” grids such as Britain and Ireland. Britain and Ireland use HVDC links to connect to neighbours, which do not provide inertia.  

Then there are so-called “lollipop” grids including Iberia, Denmark and the Baltics. Western Denmark and Iberia which have only limited AC links to larger continental systems, restricting their ability to share inertia, just to one country. Germany, France and the Netherlands, by contrast, can support high renewable penetration by sharing inertia with multiple neighbours.

When renewable production is high, so is SNSP and risk. To maintain stability, some markets curtail renewables or adjust HVDC flows to allow more spinning machines to run. In Ireland, the grid operator usually keeps SNSP below 70%, even though the formal limit is 75%, relying on flexible demand, batteries and new synchronous condensers under the DS3 programme. The chart below shows SNSP in the Irish market:

Iberia is not an island, but it has limited AC interconnection with its European neighbours. When the recent power outage hit in Spain the link with France disconnected, which instantaneously increased SNSP dramatically and may have contributed to the grid collapse. This chart shows that Iberia typically has been running at up to 70% SNSP, this is a level similar to Ireland. Since the blackout, however, the SNSP number has been reduced by the system operators. The following chart shows the situation before and after the blackout. Including forward forecast for Iberian renewable curtailment (the red peaks): 

Britain’s grid can sometimes go beyond 70% SNSP. This is possible thanks to many synchronous condensers and very fast-acting batteries, which help keep the lights on. Britain generally operates on an inertia threshold that matches this SNSP ceiling, as seen in this graphic:

Denmark is split: its DK1 zone connects to Germany, DK2 to Sweden, both via single links. Denmark was an early adopter of synchronous condensers to help manage its high renewable share, especially during periods of heavy imports or renewable output. 

Pushing the limits? 

So, can grids push renewable penetration above 70–75% where interconnection is limited? European experience suggests this is the current best practice. But that doesn’t mean progress must stall. With shared expertise, flexible markets and investment in innovations like “grid-forming” inverters, batteries and synchronous condensers, there’s no reason the transition can’t go further. 

This challenge is, at its core, an engineering problem. There are plenty of engineers ready to tackle it, and many solutions available – it’s just a matter of deploying them at scale, and quickly, to keep the net-zero transition on track. 

The electricity grids built in the 1950s, ’60s and ’70s were modern marvels. Transforming them for a renewable-dominated future in the 2020s and ’30s will be hard work, but it’s far from impossible – and the rewards will be worth the effort.

This article originally appeared as a column on www.montelnews.com