The evidence of climate change is becoming apparent in the UK, particularly in the increasing frequency of “extreme” weather events. These events differ significantly from normal patterns, are associated with severe impacts and are historically infrequent. A number of extreme weather events experienced in the last decade in the UK have been attributed to climate change including include floods, heatwaves and droughts.
The effect of these events on electricity system operation is currently unclear. The magnitude of the impact of these events may be exacerbated by factors including aging infrastructure and an increasing reliance on weather-dependent renewable energy sources as we move to a net zero energy system. This may lead to higher operational costs and complexities for system operation.
Objectives
This project will evaluate the impacts of extreme weather events on system operation up to 2050 and produce a map demonstrating the risks, probabilities, and consequences of such events at a 25km grid level of GB.
Learnings
Outcomes
It is important to understand that climate projections are used to identify changes in frequency or likelihood due to climate change signals. Climate projections continue to evolve to consider further inputs and relationships and to model intensity. The projections can indicate that parts of the GB system will increase an increasing likelihood of extreme weather events. The projections do not indicate that a specific fault will occur in a grid square in a specific year.
Through the statistical network models the project has identified a potential 4.3% increase in likelihood/frequency of transient overhead line faults by 2050. This increased likelihood of single circuit fault should be assessed further in power system models to identify whether there is an associated risk to the system or increased likelihood of coincident faults.
Alongside this transient overhead line increase a likelihood increase of 23% was identified for ‘other equipment’ by 2050. The ‘other equipment’ category includes substation equipment such as busbars, switches and cables. The number of faults in the historic fault data is small in comparison to the volume of equipment in this category on the network. The low volume of recorded faults lead to this equipment being grouped for analysis. Further investigation is required to identify the causes of this increase further and to identify whether maintenance regimes or specific asset families affect this figure.
For both OHL transient and Other Equipment cases the physical and statistical network models identified similar geographical concentrations. These faults were identified as being concentrated in the Scottish central belt, along the east coast of Scotland towards Aberdeen and along the south coast of Wales.
The demand models have identified an increasing likelihood of high temperature days, in particular in Scotland. Further investigation is required into the relationship between increasing frequency of high temperature days and an increase in cooling demand from air conditioning. There are social and economic factors to consider in assessing the uptake of air conditioning or similar technologies that would determine the volumes installed.
In the compound risks section the project has identified that high temperature events could cause a combined effect of reduced generation effectiveness, transmission network rating reductions and increased cooling demand. This scenario is also more likely to occur in late spring, summer, early autumn which is also the peak of the network outage season which will further reduce network capacity. A credible range of cooling demand increase requires further modelling. This extreme event should be considered and evaluated in more detail through the Future Energy Scenarios process and Network Options Assessment and Electricity Ten Year Statement.
During the course of the MIVOR project an extreme low temperature weather event occurred in Texas, USA. This event showed the sensitivities of Power Systems to extreme weather events. The MIVOR project undertook screening of the hazards demonstrated in this incident. It was concluded that there are similarities that could be drawn between the generation mix and networks in Texas and GB. Crucially there are differences between the systems including interconnection to neighbouring regions, the design specifications used in the more variable GB climate and differing insulation and heating methods of properties. Further consideration to increasing frequency of extreme events should be factored into future margin assessment.
Lessons Learnt
A fundamental step of MIVOR is to extract correlations and patterns from energy, network and climate data. Complete and accurate data are therefore crucial. The team has been working hard to collect, clean and integrate all the required data. Some of the findings during the collection of the data are summarised below.
- Available datasets are dispersed over multiple organizations and projects. Knowing where to find the correct datasets requires significant research. It would be of high value to catalogue, align and integrate these datasets in line with the Energy Data Taskforce recommendations. This would further enhance accessibility and transparency.
- Some valuable data is not in a consolidated machine readable format. The MIVOR project is interested in network failure data but this is recorded in a mixture of paper and digital logs, post event reports and a database of initial fault records. It would be useful to future projects to align fault data with weather conditions at the time and additional information about system conditions.
- Geographical network data is in different formats for each network operator with different meta data in each. At the boundaries between network operators some circuits are duplicated requiring additional data cleaning and processing.
- UKCP 18 projections have been relatively easy to obtain
