As the impacts of climate change continue, extreme heat events are set to move from being an anomalous edge case to a defining feature of how electricity networks operate. The UK heatwaves of 2022 and 2025, which contributed to transformer failures and outages, were early signals of a system under stress from assumptions that no longer hold.
This represents a structural shift , with higher sustained temperatures, longer heatwave durations, and reduced overnight cooling reshaping how distribution assets perform and how risk accumulates across the network. Together, these conditions are reshaping how distribution assets perform , and how risk accumulates across the network.
Extreme heat is changing the operating envelope
Electricity networks are particularly exposed. Transformers are designed around ambient temperature assumptions that are now regularly exceeded. As temperatures rise, cooling efficiency declines, insulation ageing accelerates, and operational margins narrow, all while demand continues to increase, eroding headroom precisely when assets are most vulnerable.
The hidden risk sits inside substations
When assessing climate risk, focus typically falls on outdoor temperatures. Yet one of the most significant vulnerabilities sits inside substations, which were not designed as thermally managed environments. Instead, they often rely on passive ventilation, minimal solar protection, and construction optimised for cost and safety rather than heat performance. This has given rise to building envelopes that, under extreme heat, can drive indoor temperatures above ambient levels.
Internal substation environments can be materially hotter than external conditions, creating hidden, unmanaged thermal stress on critical assets. Drivers include:
- Thermal mass storing and releasing heat
- Solar gain on masonry structures
- Urban heat island effects
- Limited airflow in enclosed or underground sites
Transformers in brick, stone or integrated substations are more likely to experience elevated hotspot temperatures and accelerated ageing. In turn, this creates broader operational challenges, including safety risks above 40 °C, reduced maintenance windows, and diminished flexibility during periods of peak stress.
A 6 °C rise in winding hotspot temperature can double insulation ageing rates
At 120 °C, transformer life can fall from decades to ~2.2 years
At the system level, ambient assumptions are already being exceeded:
Assets designed around historic ambient temperature assumptions are increasingly exposed to heatwave conditions of 35–39°C,with even higher temperatures possible in enclosed or thermally constrained environments.
Urban areas can run 4–5 °C hotter than surrounding regions
Thermal stress is not linear — small temperature increases can trigger step changes in asset life, capacity and failure risk. Operationally, this translates into immediate constraints:
~1.5% capacity loss per °C (self-cooled transformers)
~1.0% per °C (forced-air systems)
Elevated ambient temperatures reduce the thermal rating of transformers, cables and other network assets, decreasing available network capacity during heatwave conditions. The extent of this reduction depends on asset type, loading, cooling arrangements and local environmental conditions.
The bigger risk: invisible misallocation of capital
The most significant strategic risk is not just overheating — it is a lack of visibility. Without granular thermal data, networks are forced into reactive investment decisions, and ultimately the misallocation of capital.
Without visibility, networks risk solving the wrong problem in the wrong places — at scale. Some assets are over-protected. Others remain exposed. This transforms extreme heat into a risk management and investment problem, not just an engineering one.
Five principles for heat-resilient networks
The following principles have been identified as being central to the transition to more heat-resilient networks:
1. Visibility before intervention
Thermal risk must be measured, not assumed. By monitoring indoor substation conditions alongside asset-level thermal performance, operators can make more targeted, cost-effective decisions.
2. Passive first, active last
Passive measures consistently deliver the highest return — from ventilation optimisation and shading to reflective materials and improved internal layouts. Active cooling also has a role, but is likely to be limited to extreme cases only.
3. Design standards must evolve
Future-ready design should account for higher ambient temperatures, microclimatic effects, orientation and airflow, and material performance. Decisions made at the design stage represent the lowest-cost opportunity to embed climate resilience at scale.
4. Dynamic operation, not static ratings
Static ratings do not reflect reality or allow for flexibility. Dynamic approaches enable greater utilisation under normal conditions while providing protection during extreme events, improving both resilience and overall efficiency.
5. Learning from hotter climates
Established practices in warmer regions prioritise passive design, monitoring-led intervention, and targeted upgrades, while notably avoiding over-reliance on energy-intensive cooling.
Passive measures consistently deliver the highest return — from ventilation optimisation and shading to reflective materials and improved internal layouts. Active cooling also has a role, but is likely to be limited to extreme cases only.
3. Design standards must evolve
Future-ready design should account for higher ambient temperatures, microclimatic effects, orientation and airflow, and material performance. Decisions made at the design stage represent the lowest-cost opportunity to embed climate resilience at scale.
4. Dynamic operation, not static ratings
Static ratings do not reflect reality or allow for flexibility. Dynamic approaches enable greater utilisation under normal conditions while providing protection during extreme events, improving both resilience and overall efficiency.
5. Learning from hotter climates
Established practices in warmer regions prioritise passive design, monitoring-led intervention, and targeted upgrades, while notably avoiding over-reliance on energy-intensive cooling.
Static ratings do not reflect reality or allow for flexibility. Dynamic approaches enable greater utilisation under normal conditions while providing protection during extreme events, improving both resilience and overall efficiency.
5. Learning from hotter climates
Established practices in warmer regions prioritise passive design, monitoring-led intervention, and targeted upgrades, while notably avoiding over-reliance on energy-intensive cooling.
Network Implications
Extreme heat is likely to become an increasingly significant constraint on electricity networks – shaping asset performance, network capacity, investment decisions and regulatory outcomes.
For stakeholders, the implications are immediate.Operators must shift from reactive mitigation to proactive resilience, investors must price thermal risk into long-term value, regulators must align standards with future climate realities, and innovators must focus on scalable, deployable solutions.
The tools to respond already exist. The gap is no longer technical — it is strategic. The constraint is no longer what we know — but how quickly we act on it.
This requires a shift from incremental adaptation to coordinated action:
- Embedding climate assumptions into design standards
- Making thermal visibility a core operational metric
- Scaling proven solutions beyond pilots
- Aligning incentives with long-term resilience
This article is informed by ongoing collaborative work between PNDC and UK Power Networks, aimed at translating technical insight into targeted, strategic investment priorities for more resilient electricity networks.
To discuss the findings or explore opportunities for collaboration, please get in touch. You can also connect directly on LinkedIn with PNDC’s experts in this area:
Dr Priya Bhagavathy, Edward Corr, Dr Callum Rae, Dr Samir Soares.
