What it takes to manage the power grid in a heatwave

Posted: August 18, 2026

As successive heatwaves swept across Europe this summer, the power system struggled. In June, French nuclear reactors were forced to shut down because their cooling water was too hot to be discharged back into rivers. Two months later, a prolonged drought lowered parts of the Danube to record levels, threatening to shut down the power plant that produces nearly half of Hungary’s entire electricity.

At the same time, low wind—a common companion to high temperatures—have slowed down wind turbines. And power lines sagged in the punishing sun, taking their load capacity down with them. Making matters worse, people switched on their air conditioners all over the continent, causing demand to spike.

Similar heatwaves have strained the grid in recent summers, not just in Europe, and are bound to keep intensifying. Far from an isolated incident, the last few months illustrate how much more demanding power system operations have become in this changing climate. In response, grid operators are becoming more creative in how they balance the system and ensure electricity continues to flow—hardening their infrastructure, actively managing demand and trying to keep on top of all the extra data they collect to do so.

“Complexity in the electricity grids is not new,” says Anna Lafoyiannis, an expert in transmission operations at the Electric Power Research Institute who previously worked in reliability assurance at Canada’s Independent Electricity System Operator. “That means there’s a lot of experience in dealing with that. But it is increasingly coming to a head.” 

Hardening and enhancing the physical network

The first line of defense in a heatwave is the grid’s physical health. Utilities are now performing vulnerability assessments on each individual asset to manage their risk and guide investment decisions. Part of this work is guided by standards bodies and regulators, who want companies to be prepared for a wider range of high-risk events.  

Advanced planning and proactive response is becoming more important as companies deal with increased complexity and compound events, according to Lafoyiannis, since extreme heat rarely comes in isolation. “It’s never just a heatwave. It’s a heatwave and the wind isn’t blowing, or it’s a heatwave and there’s a tornado,” Lafoyiannis says.

Grid operators now have a suite of tools at their disposal to harden their infrastructure. One is reconductoring: replacing traditional transmission lines with advanced conductors that can carry up to 40% more load. They also sag far less in high temperatures, which reduces the risk of lines coming into contact with vegetation to spark wildfires. Utilities have also long used covered conductors, which are insulated by an extra outer layer to avoid faults and fires; Southern California Edison has installed thousands of miles of these in wildfire zones. In the U.S., some companies are now even tracking insulator faults hundreds of miles downwind of fires, where conductive soot can settle on the insulator and spark additional blazes.

Another increasingly popular measure is adopting more granular real-time monitoring of the network’s actual state. Historically, network capacity was managed according to static line ratings—fixed assumptions based on seasonal worst-case weather scenarios, which dictated how much current could be carried without overheating or dangerous sag. The Federal Energy Regulatory Commission in the U.S. now mandates that network operators instead use ambient adjusted ratings (AAR). These use forecasts for ambient air temperature, allowing operators to raise capacity when conditions are favorable. In March, PJM became the first regional grid operator to implement AAR, and its ratings now adjust hourly across 47 regional weather forecast zones.

Going one step further, some networks have started trialing dynamic line ratings, installing sensors that rate transmission capacity based on real-time readings of local temperature, wind speed and solar irradiation. This allows operators to use their assets at maximum efficiency. As Lafoyiannis puts it: “When the wind cools a line, you can push more power through.” 

Some utilities in the Midwest have experimented with dynamic line ratings, reporting capacity gains of up to 40%. Network operators in Europe have also installed the sensors, although the technology is not yet widespread. That could soon change, though. Lafoyiannis says EPRI, which runs an outdoor test line for dynamic line rating systems, is just wrapping up its first pilot to test performance of six different vendors. It is now planning a second round because of growing interest and availability. Lafoyiannis says there are now over 40 companies, each with their own “special sauce.” Some are even sensor-less, relying on third-party weather data, advanced thermal modeling and AI-driven analytics.  

Michael Dodd, an energy networks expert at assurance and risk consultancy DNV, says the system doesn’t just provide relief when supply is restricted and demand spikes during a heatwave, but can also save costly network expansions. “It means that you're freeing up capacity without having to build necessarily as much as you otherwise would do,” Dodd says.

Managing soaring power demand during heatwaves

Even with a network in top shape, high demand during a heatwave can strain the system. This is why most utilities now resort to actually lowering demand itself during the hours when air conditioners are humming but power plants may be compromised. “The cheapest megawatt that you can use is the one that you don't have to use—the one that’s already there,” Dodd says. “Increasingly, system operators are trying to tap into that demand side.” 

There are several ways to do this. In California, SCE uses time-of-use residential rates to incentivize its customers to lower consumption at critical periods. Prices rise during peak afternoon and early evening hours, when demand usually spikes in the summer months. In the U.K., the National Energy System Operator recently tried paying customers to turn up or down their energy use at certain times, and to plug in their electric vehicles to allow the grid to draw extra power.

Operators are also implementing demand response for commercial and industrial consumers. During a recent heatwave in Toronto, where Lafoyiannis lives, the Independent Electricity System Operator relied on a whole mix of demand management—adjusting hundreds of thousands of smart thermostats, switching off charging for EVs and curtailing industrial power consumption.

Industrial demand response and virtual power plants—the technical term for those aggregated thermostats—are likely to grow in importance as demand increases with the growth of data centers, the electrification of industry and, in areas like Europe, a proliferation of air conditioners.

How utility control rooms are evolving to cope with extreme heat

All this naturally poses another challenge: managing the vast amounts of data from grid sensors, real-time forecasts and the signals from a suddenly far more expansive fleet of grid assets. Grid operators’ control rooms “now have far more real-time information that they're able to take advantage of to support the decisions that are being made,” Dodd says. “So the back offices and support functions of those control rooms are changing rapidly.” 

For example, Lafoyiannis says operators are beefing up their weather monitoring with in-house meteorologists. They are also using multiple external forecasters—so they don’t have one singular snapshot, but a more probabilistic prediction. That comes in handy for evaluating physical risks to the network. With a better idea of what’s coming, operators can also direct maintenance crews and protect vulnerable infrastructure during extreme weather. And it helps with predicting when solar production will peak and wind turbines will spin.

European utilities, in particular, have an edge when it comes to boosting situational awareness, Lafoyiannis says. Some are now using AI and machine learning to improve their forecasts. (The same goes for long-range climate modeling, which can help decide where grid upgrades will be needed.) According to Dodd, utilities are also turning to more complex digital twins to mirror what's happening in the broader energy system. It’s no surprise that one of the biggest areas of growth for DNV is assuring that the data streams these companies are feeding into their control rooms are as robust as possible.

It’s easy to look at all this and see a new frontier in data management. But, as Lafoyiannis, points out, the power sector was an early adopter of artificial neural networks, turning to AI for load forecasting as early as the late 1980s. “This is an industry that will adapt to technology and will use big data,” Lafoyiannis says. “I think that’s in our DNA.”