Salford University will next month begin a three-year European project to develop software that learns from a building's energy and adapts consumption to reduce carbon emissions.
The Seeds project (Self learning Energy Efficient builDings and open Spaces) is funded by the European Commission and contains nine partner universities and companies. The project aims to develop wireless sensor technology and data mining methods that learn to improve energy consumption whilst maintaining user comfort.
Prof Sunil Vadera from the school of computing, science and engineering, is leading Salford's £299,000 section of the £2.2m three-year project. He said: "This is a project of major importance as it brings together scientists from different specialisms and locations.
"Using software in this way has the potential to make our buildings much more energy efficient without having to rely on every single person using them in the correct manner."
At Salford, computer scientists will have responsibility for applying self-learning software to the system which will allow the buildings to become more efficient over time as data is gathered by sensors. The university will also lead on work to encourage uptake of the new technology using knowledge developed at its Energy Hub to change people's poor environmental behaviours.
The project's first demonstration sites are part of the University of Stavanger campus in Norway and an office block in Madrid. They were chosen because of their high levels of energy use and contrasting local lifestyles and weather conditions.
In practice this will involve collecting various data, such as temperature, humidity, luminance, and occupancy via wireless sensors. The software then learns to optimise heating and ventilation so that user comfort is satisfied but energy consumption is minimised.
Energy savings should follow since the heating or air conditioning will be controlled more closely to actual requirements such as avoiding being on when there are no people in rooms.