13 November 2023
By Giulia Cirri
It is no news that increasing temperatures will be a continuous problem across the world. July 2023 was the hottest month ever recorded by humans, with the average global temperature at 1.5C above pre-industrial levels. The UK will be no exception. Even if after this summer it is harder to imagine, UK summers are increasingly getting hotter, for longer, with more acute and more frequent heat waves.
There are three facts we just can’t ignore. I’ll keep it brief:
1. Temperatures are increasing.
2022 was the warmest year for the UK since records began in 1884. All the 10 warmest years have occurred since 2003.
2. Higher temperatures are here to stay.
In fact, they are only going to get worse in the short run. Even with decreasing emissions, global warming would likely continue to happen for a few years as emissions persist in the atmosphere. This is not to say that efforts to combat climate change are useless. In fact, every tenth of a degree matters. What it does mean is that we will have to get used to more extreme weather events more frequently, from heat waves to extreme cold.
3. The UK is so, so, unprepared for a change in temperature.
We have all heard it before, and more than that, we have all experienced it. The UK housing stock is massively underperforming in terms of heat-proofing, the railway system collapses at the first sign of heat waves, and in London, seven tube lines do not have any air conditioning. Last year alone, there have been 2,800 more deaths among people aged 65 in England during the heatwaves. As for most things, the harshest effects of climate change are felt more strongly by the most vulnerable parts of society.
UK homes are not fit to be safe places during the summer. As we’ve seen this winter, they’re not fit to be safe places in the winter either. Heat insulation works year-round: the same measures that will make your house retain heat in the winter will also keep heat out in the summer, as insulation is basically slowing down the transfer of temperature between the exterior and interior of your house. This is not just a matter of climate adaptation. A building less able to maintain its internal temperature will require more energy to be kept cool in the summer or hot in the winter, increasing the buildings’ direct or indirect greenhouse gas emissions. Basically, a poorly insulated house suffers more from the impacts of climate change but also contributes more to climate change.
It is not only about the climate, either. Higher energy requirements due to low insulation directly translate into higher energy bills. There are health and social costs that come with an inadequate building stock, both in the summer and in the winter. Energy efficiency is a driver of fuel poverty, which refers to households that, when they spend the required amount to heat their home, are left with a residual income below the official poverty line. In turn, fuel poverty has serious impacts on physical and mental health, including heart attacks, strokes, and social isolation, particularly for the most vulnerable sections of society. As mentioned, last year, there have been 2,800 more deaths among people aged 65 in England during the heatwaves.
In short, the condition of UK houses is a problem that concerns us all and touches several areas of our lives. So, what is the size of the problem in the UK?
According to the 2021 UK Heat and Buildings Strategy, there are about 30 million buildings in the UK. In total, these buildings are responsible for around 30% of our national emissions, with the vast majority of these emissions resulting from heating: 79% of buildings emissions and about 23% of all UK emissions.
These emissions are only in part due to low insulation. Lowering the environmental impact of buildings also includes changing the energy and heating sources and moving to renewable energy or more sustainable heating methods at the level of the individual home and at the grid level.
The easiest way to understand the energy efficiency of UK buildings is to look at Energy Performance Certificates (EPCs). EPCs have been mandatory since 2007 for almost all residential buildings in the UK. These standardised certificates are compiled by an energy assessor who collects information on a building’s characteristics, rates its performance, and estimates consumption and cost. Energy efficiency is scored out of 100 and assigned a label rating the building on a scale from A to G, with A being the most efficient rating. EPCs also provide the building’s energy efficiency potential, recommendations on the changes to make to improve energy efficiency, and their cost. Importantly, EPCs scores are not influenced by the occupants’ usage behaviour but are purely based on information about the building. The average EPCs in the UK is D, and out of 30 million buildings in the UK, around 19 million homes have an EPC rating of D or worse. [by the way, all EPCs are publicly available here].
Both reducing fuel poverty and improving the stock of UK buildings have been on the to-do list of UK governments for a while – most recently, it was the 2021 Sustainable warmth and the UK Heat and Building Strategy. The latter, in line with the 2017 Clean Growth Strategy, set a goal to have the 2.5 million fuel poor homes in England improved to an energy efficiency rating of C or better by 2030. More broadly, the aim is to have as many homes as possible improved to EPC Band C by 2030.
The achievement of these goals hinges on one keyword: retrofitting. Retrofitting indicates the changes made to an existing building (through new materials, products, or technologies) to reduce the energy needed to occupy that building. It relates to those changes that are specifically designed to affect the building’s energy performance and not, for instance, increase its aesthetic value.
Retrofitting measures can be broadly divided into two groups. The first includes measures to increase insulation, such as insulating roofs, walls and floors, or applying double or triple glazed windows or draft exclusions to doors. The second includes reducing buildings’ energy consumption, for instance, swapping boilers for low carbon heating/cooling systems (solar, micro-wind, heat pumps, green hydrogen etc.). In most cases, the gold standard requires applying insulation measures first, so that the new heating or cooling technologies can benefit from improved energy efficiency. Heat pumps, for instance, are considered to be the main method to decarbonise buildings, but in order to work properly in most cases they need to be installed in well-insulated homes.
These technologies, and using them for “smart buildings”, are usually the first thing that comes to mind when we think about the use of AI for energy efficiency. Smart buildings are buildings in which their core systems are linked through various smart environmental control technologies, occupancy sensors, and building management systems to share information and automate various processes, such as heating, ventilation, lighting, and air conditioning. Smart buildings do not necessarily use AI, but AI can make them smarter. Indeed, these sensors produce large, continuous quantities of raw data, and, in short, AI is very good at making sense of large quantities of raw data and finding optimisation strategies, reducing energy use. Projects like RESPIRA, are a good example to showcase the potential of AI for smart buildings. RESPIRA, implemented in the Barcelona metro network, balances temperature, humidity, energy consumption and air quality to predict the environmental and operating conditions of the facility, reducing overall energy use (and operating costs) through efficient climate control.
While these whole-systems approaches are fascinating, they are more likely to be implemented in public buildings, such as hospitals, or schools. The average UK home is unlikely to turn into a fully smart building in the short term. The idea of smart, optimised buildings is exciting, and smart buildings have become a bit of a buzzword in urban design. Nevertheless, there is a risk that this fame will result in smart buildings being perceived as the only use of AI for energy efficiency, fostering the image of AI as something reserved for sci-fi like buildings reserved to the likes of Gwyneth Paltrow or Elon Musk. Moreover, it can’t be ignored how smart buildings have been raising more than an eyebrow on the issue of data privacy, as they rely on massive amounts of data that can be considered sensible.
Therefore, I was curious to know if there are other, maybe less glossy, ways in which AI is currently being used to support improvements in energy efficiency. In particular, I was curious to investigate the use of AI to improve the physical building stock, which the UK desperately needs to do to raise EPCs ratings. In the next blog, we’ll look at novel uses of AI to predict retrofit effects through data-driven simulations and identify the best retrofitting strategies.