Universities Employ Artificial Intelligence to ‘Learn’ Best Energy Performance
Six universities across Britain are enlisting support from Energy Assets to employ Artificial Intelligence (AI) to reduce energy consumption in buildings.
Facilitated by The Energy Consortium (TEC), Anglia Ruskin, Bath, Bristol, Newcastle, Regents and York universities have all embarked upon a pilot scheme with Energy Assets’ AMRdna service to use AI, developed by kWIQly, to identify energy waste.
Energy meter data is being collected automatically every half hour from across university estates, and then interrogated by AMRdna to map the consumption profile for each building. Using this information, the AI ‘learns’ the optimum ‘model’, taking account of multiple variables, such as weather conditions, Day Light Saving Time, hours of operation etc..
Since every energy-related action, whether it’s switching on a light, or starting a heating system, leaves a trace in energy data, the AI spots anything that is outside the best possible performance profile. By analysing the data in this way, any wasted energy hiding in plain sight is identified, automatically alerting energy managers to the required corrective action.
TEC is running the trial for its university members through its Flexible Gas Framework and is collaborating with estate management teams to evaluate the benefits delivered by AMRdna.
Early results are impressive. For example, at Anglia Ruskin University, 72 building types, including lecture theatres, halls of residence and offices, form part of the project. Already, energy waste patterns have been identified at six different sites, including one hall of residence where a change in heating operation prompted by AMRdna resulted in a year-on-year consumption improvement of 38%, without any adverse impact on students.
Simon Chubb of Anglia Ruskin: “The AMRdna analysis enabled us to get a very clear understanding of the problems facing our portfolio. The ability to very quickly identify sites where consumption is over what we expect has enabled us to reduce our time to correction dramatically.”
Steve Creighton, Head of Member Services at TEC: “Consumption reflects when a plant operates. Activity can be reverse engineered from consumption allowing AI to search, quantify and prioritise. This in turn makes it possible to manage problems on a daily basis for even the largest estates. As time-to-action is critical in reducing waste, early identification and diagnosis enables financial gains and carbon reduction for our members.”
For more information contact:
Energy Assets Limited
telephone: 01506 405405 | email: firstname.lastname@example.org | website: www.energyassets.co.uk
view profile on ESTA at: www.estaenergy.org.uk/members/energy-assets-limited
All Energy Assets Limited Case Studies
- Universities Employ Artificial Intelligence to ‘Learn’ Best Energy Performance
- Six universities across Britain are enlisting support from Energy Assets to employ Artificial Intelligence (AI) to reduce energy consumption in buildings. | 6th April 2018