Researchers said on Friday, in a Nature Energy study, that France, Spain and Italy face the highest risk of heatwave power cuts among eight Western European grids, with the UK comparatively resilient.
The team, from the University of Cambridge, City University of Hong Kong and The Chinese University of Hong Kong (Shenzhen), said the UK still faces a rising risk of failure as heatwaves intensify and demand peaks.
The team examined Spain, Portugal, France, Italy, Germany, Belgium, the Netherlands and the UK, all of which have been hit by recent historic heatwaves.
They said existing models capture only parts of the problem and at high computational cost, so they built algorithms to model the full picture at lower cost.
Heatwaves raise demand for cooling while cutting supply, the researchers said. Solar panels lose efficiency as heat rises, conventional and nuclear output is reduced to prevent overheating, and the high pressure that brings hot weather often brings low winds that weaken wind farms.
Power lines must also carry less load to avoid hitting their heat limits.
Together these effects can lead to load shedding, which the researchers define as intentional power cuts to reduce demand and protect power systems.
Simulations cover 2026 to 2030
The team simulated hundreds of projected heatwave scenarios for 2026 to 2030, built from the heatwaves of 2019, 2022 and 2024, using open-access grid, weather and demand data.
The paper says the scenarios are meant to test grid operation, not to predict when or how individual future events will happen.
In the French grid, average load shedding reached 0.45 per cent under projected heatwaves. Spain and Italy also showed load shedding, while the remaining grids, including the UK and Germany, showed no substantial load shedding in the scenario set, the paper said.
Cross-border help is uneven. The paper says France gains substantial relief when interconnected with less-stressed neighbours, with load shedding falling when linked to Spain, Italy or the UK.
Spain gains minimal relief from joint simulations with Portugal or France, because neighbours under the same heat have little surplus power to send.
The researchers also tested storage.
They assumed existing storage is fully charged when a heatwave arrives, which may not always be the case.
Co-corresponding author Professor Minghua Chen of the Chinese University of Hong Kong (Shenzhen) and City University of Hong Kong said even then storage is not enough to prevent load shedding.
“Even under that ideal assumption, existing storage resources are still insufficient to prevent load shedding,” he said.
Findings for 2026 to 2030 exclude datacentre demand growth
The findings assume only a conservative annual rise in demand over the next five years and exclude the emerging boom in datacentre construction. The model keeps the transmission, generation and storage assets in public data and does not include future infrastructure expansion.
The authors also cite uncertainty in climate and demand projections and the limited resolution of publicly available grid data. They said validation with proprietary, grid-specific data would strengthen the quantitative estimates.
The paper sets out three adaptation routes: demand response in affected load centres, reconductoring or reinforcing vulnerable lines, and storage near bottleneck locations. The researchers also suggest a cooling, reflective coating for powerlines.
The data sets, algorithms and source code are being released as a downloadable toolkit called HEAT-Analysis, which operators can use by adding their own infrastructure data, the University of Cambridge said.
“We hope power providers will use these tools to identify vulnerabilities in their systems and take necessary action to stop them collapsing during future heatwaves,” said co-corresponding author Professor Srinivasan Keshav of Cambridge’s Department of Computer Science and Technology.