ENERGY IN EDUCATION
The UWTSD IQ building in Swansea
Clear benefi ts UWTSD started using the system in the latter part of 2023 and has already seen some significant benefits. For example, when heating had been inadvertently left on in an unattended building, the machine learning system flagged this, and it was fixed almost immediately. “Of course, some spikes are easily
IMAGES: UNIVERSITY OF WALES TRINITY SAINT DAVID
Transforming sustainability in higher education
The University of Wales Trinity Saint David (UWTSD) is investing in machine learning technology from Energy Assets to support its ambitious plans to be a leading university for energy sustainability.
T
he UWTSD, which has campuses in Swansea, Carmarthen, Lampeter, Cardiff, London and
Birmingham, is already reaping the benefits of a solar power strategy – now it’s turning to artificial intelligence (AI) to transform energy efficiency across its entire estate. The University is using AMR DNA,
an Energy Assets service, to apply AI- informed machine learning analytics to drive out energy waste, optimise efficiency, reduce consumption and make significant steps toward net zero carbon emissions. The AMR DNA software, powered by kWIQly, progressively learns what the best energy performance looks like in each building and automatically flags up in near real-time unusual spikes in energy usage, which can be quickly addressed by the University’s management team. “Our vision is to build on our existing position as a leading UK university for energy sustainability,” says Dan Priddy, finance and business performance manager at UWTSD. “We’re taking real actions that will reduce Scope 1 (direct) and Scope 2 (indirect) greenhouse gas emissions by 95% by 2030. “This will be made possible by
infrastructure upgrades, more investment in renewable energy, the adoption of net zero construction standards on new buildings and, critically, by applying machine learning technologies. “We’re already seeing the benefits of this approach. Thanks to the installation of solar panels, the electricity demand at the Dynevor building, home to our Swansea Art College, is significantly met by renewable generation. Now, with the move to machine learning, we aim
24
explained,” says Dan, “but where there is nothing obvious, we see the AMR DNA alert and ask the site teams to investigate. It means we have a clearer picture of energy consumption across the entire estate, with a greater ability not only to spot consumption spikes more easily but also to address them quickly.”
AMR DNA currently tracks
performance across gas services but is about to be trialled at UWTSD to monitor efficiency in electricity networks, using data collected from fiscal meters and sub-metered areas.
Students are encouraged to engage with the University’s sustainability ambitions
to ensure that our entire estate is optimised for energy efficiency and emissions reduction.”
Identifying consumption Working with the University, AMR DNA analysed a year's worth of historical meter data to identify the unique ‘fingerprints’ of consumption that would provide an energy benchmark for each of the UWTSD buildings. The software models what ‘normal’ consumption looks like, taking account of multiple factors, such as occupancy levels and operating hours, and ‘learns’ what
optimal performance should look like. Using pattern recognition linked
to key performance indicators, the system interrogates metered data to spot tell-tale signs of energy waste. This waste can result from something as simple as equipment running needlessly or lights being left on overnight, or might be linked to incorrect heating time clock controls, or over-compensation for ambient weather conditions. “We operate six campuses
covering everything from student accommodation and teaching spaces to engineering workshops, sports facilities,” explains Dan. “This creates an estate with a lot of nuances, which makes the manual tracking of energy consumption very difficult and time- consuming.
“But with machine learning, we not
The university expects to generate 700,000 kWh of solar power in one year
only see the big trends, but we also pick up on smaller issues that we would never have identified in the grand scheme of things…but they all add up. The amount of manual resource you would need to capture this level of detail would be ridiculous, but now we can, using AI.”
Engaging with students UWTSD sets its energy efficiency programme in a wider sustainability context and is active in engaging with its student population, running induction programmes for new students and staff and holding an annual Sustainability Week, while joint action with the Students’ Union has resulted in significant progress in waste reduction and recycling rates, along with improvements to the local ecosystem by cultivating wildflower meadows. Students on environmental and sustainability courses are engaged in practical projects using the University estate and this includes using energy data from the University in real-life case studies. The University has recently established an Energy Efficiency Group dedicated to spearheading efforts towards consumption reduction within campuses. This aims to serve as a focal point for coordinating, implementing, and monitoring various energy-saving initiatives, to build accountability, ownership and to drive associated behavioural change across staff and students. Significant infrastructure upgrades
are also underway, notably via Salix loan support for renewables, in smart metering, electric heating, double and triple-glazed replacement windows and LED lighting. In the next 12 months, UWTSD expects to generate a total of 700,000 kilowatt hours of onsite solar power – satisfying about 12% of its total anticipated electricity demand. The University’s remaining electricity needs are all sourced from
zero-carbon providers. ■
www.energyassets.co.uk
EIBI | FEBRUARY 2025
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
