This research project involves the design and development of a sensing infrastructure that consists of networked physical (e.g. presence sensors, power consumption sensors) and virtual (e.g. calendar and room booking sensors, application usage sensors) sensors, which will provide fine-grained information about how much energy is being used, for what purpose and by whom. By applying techniques from knowledge engineering, activity recognition and machine learning, the project involves the use of high level information in order to link usage patterns to real-world activities and workflows.
This information is used to parameterise building models used in building management to more accurately predict energy usage and to optimise (decentralised) energy consumption, generation and storage and to assist in the development of a decision support tool that visualises the collected data as well as the expected impact of energy saving strategies such as organisational changes and policies or the rescheduling of activities. This will enable decision makers to identify where energy is being wasted and to formulate and evaluate strategies to reduce energy consumption. The data will enable a better understanding of the way the building is used and how heat is wasted. Through a combination of physical and virtual sensors, a more accurate measurement of thermal comfort of the building’s occupants will be established and thus assist in resolving ever occurring complaints and potential conflicts associated with the diverse needs for occupant comfort in buildings that can result in unnecessary overheating.
Date: Sept 2010 – Sept 2013
Funding: EPSRC (Engineering and Physical Sciences Research Council) £606,679
Researchers: Tony Roskilly (PI) – Sir Joseph Swan Centre for Research on Energy. Patrick Olivier, and Thomas Ploetz (CIs).
Dan Jackson, Cas Ladha, Karim Ladha, Aftab Khan, Sam Mitchell Finnigan, Chris Kray, Mohammad Royapoor, Zhiwei Gao – Sir Joseph Swan Institute for Energy Research.