One of the many collaborations Imperial College London has with universities around the UK is the EPSRC-funded Centre for Doctoral Training in Future Power Networks & Smart Grids. It brings together the College’s Department of Electrical and Electronic Engineering with its counterpart at the University of Strathclyde. One of the students, Magnus Jamieson, who is currently based up North, has very kindly written us a piece on reliability in the energy sector in the face of climate change.

Weather, the UK has a lot of it and it’s a major factor in interruptions to our energy supply. As climate change makes extreme weather events more frequent, and more intense, reliability of our energy sector becomes an even more pressing concern. Power companies have to continuously improve performance while also keeping their costs, and our bills, down. My research focusses on knowing what is causing outages, where, and how to address them, challenges that urgently need addressing.

Currently the fundamental role of the power sector is to get energy from a source and convert it to electricity, before then delivering it to customers. However, people don’t use “electricity”, they use the services and technologies made available by electricity; phones, televisions and lights. It is not the loss of electricity that affects people but the loss of these services. When these services are needed most is also when the power system is at its most vulnerable, during storms and inclement weather. It is important that a power system is designed to be resilient in the face of extreme weather conditions.

All of these weather conditions have three aspects in common, uncertainty in weather forecasting, the interdependency of the power system with other infrastructures and changes in frequency due to climate change. In order to reduce interruptions it is necessary to know the significance of each fault source, and how the frequency of these faults will change. As this will guide investment in networks improved techniques for modelling the effects of weather on network fault rates have to be investigated.

Among the major challenges for resilience and reliability improvements in the UK is the fact that there is so much aging infrastructure.  There are assets on the network which have been operational since the early 20^th century, and many cannot be relocated because of the challenges associated with getting planning permission for energy infrastructure, particularly in environmentally sensitive areas, where flooding can be a high risk. In 2015 an old power station in North West England, which had been converted to a substation in the 1950s, was flooded affecting thousands of people. During its conversion it was deemed that to move it elsewhere for flood mitigation would be “extremely difficult” due to planning permission complications.

The nature of the UK land and property market also means that land is used sub-optimally. Phenomena like “land banking” contribute to an economy whereby land is used poorly from a strategic point of view or its ownership incredibly concentrated. This affects infrastructure development by artificially inflating the cost of land, making redevelopment or retrofitting preferable to relocation in many cases.

Evidence from Hurricane Katrina also suggests that network resilience is more significantly affected by a large number of smaller disturbances as opposed to losing a small number of major infrastructure assets. For example malfunctioning switches or fuses over a wide area is a bigger problem that a substation failing. Improving many smaller network assets instead of carrying out major work on fewer, larger assets in itself carries risk and emphasises the need for accurate projections.

In order to keep costs down whilst still maintaining a secure and reliable power system we have to develop tools which account not only for the inherent faults possible in networks due to random malfunction, but also to account for high-impact-low-probability events such as flash floods or major blizzards. We also have to account for how these weather patterns will change due to climate change. Some weather types are likely to become less frequent– most notably snow and cold related faults – but others may become more frequent and intense, such as flooding. Knowing where to direct investment for the current network is also dependent on factors such as storage, distributed, small-scale generation, and the types of generation connected to the main transmission networks.

There are a plurality of issues facing the power sector, only exacerbated by changeable weather conditions. The challenge of how we quantify reliability and resilience in the face of extreme weather conditions is at the forefront for the energy sector. My research will help fill in some of the gaps but we needs more researchers to not only develop these models but ensure their accuracy and versatility as networks and weather patterns change.