Analysing occupant behaviour in buildings to reduce energy use

PhD researcher Rima Alaaeddine was selected to present her research at Westminster

ARTIFICAL intelligence has been at the centre of many media reports recently and a researcher at the University of Huddersfield will demonstrate how it can be a better way to better predict building energy usage. 

Rima Alaaeddine is a PhD researcher within the University’s School of Art Design and Architecture.  Her research aims to combat and minimise the ‘energy performance gap’ and could benefit the building sector at a time when there is increasing pressure on industries around the world to conserve their energy consumption. 

The term ‘energy performance gap’ arises when a building consumes more energy than was initially predicted during the design phase.  This gap is attributed to a set of variables such as environmental conditions, building characteristics and occupancy.

Occupants have significant impacts on building energy use, and there is complexity in predicting how much energy a building’s occupants will consume and the way they individually interact with the building on a daily basis, known as ‘Occupants Behaviour’. 

This includes actions such as their use of lighting, hot water, electricity, appliances and the way they interact with the building for example, opening windows and controlling their heating, ventilation and air conditioning systems. 

Rima’s research could play an important part in helping the construction industry meet strict energy efficiency targets, recently set by the UK Government as part of a new energy strategy.  With the energy consumption of buildings accounting for 30 per cent of the entire global energy use, improving the energy efficiency of buildings is one of the key strategic objectives.  More accurate energy predictions can facilitate building energy optimisation and guide decisions regarding the building energy performance. 

“My research will employ a branch of Artificial Intelligence (AI) entitled Machine Learning,” said Rima.   

She explains that by employing machine learning techniques, capable of handling complex and non-linear problems, AI can offer more accurate predictions on occupants’ behaviour.

Rima’s project is already receiving national recognition.  The 27-year-old researcher was shortlisted from hundreds of applicants from across the UK to present her research in Parliament, as part of the annual STEM for BRITAIN competition, to a range of politicians and a panel of expert judges.

The prestigious poster competition, headed by the Parliamentary and Scientific Committee, was organised in collaboration with the Royal Academy of Engineering, the Royal Society of Chemistry, the Institute of Physics, the Royal Society of Biology, The Physiological Society and the Council for the Mathematical Sciences. 

Rima’s entry was entitled ‘Minimizing the energy performance gap by application of an integrative machine learning methodology for occupants behaviour prediction’ and she said it was an honour taking part in STEM for BRITAIN and to be given the opportunity to present her work in Parliament.  

“The event provided me with an opportunity to communicate my research as widely as possible, to inform and enthuse non-scientific audiences about my research in the building energy performance realm aiming to unveil the benefits it brings,” she said.