We study the problem of designing sensor sets for capturing energy events in buildings. In addition to direct energy sensing methods, e.g. electricity and gas, it is often desirable to monitor energy use and occupant activity through other sensors such as temperature and motion. However, practical constraints such as cost and deployment requirements can limit the choice, quantity and quality of sensors that can be distributed within each building, especially for large-scale deployments. In this paper, we present an approach to select a set of sensors for capturing energy events, using a measure of each candidate sensor's ability to predict energy events within a building. We use constrained optimisation -specifically, a bounded knapsack problem (BKP) -to choose the best sensors for the set given each sensor's predictive value and specified cost constraints. We present the results from a field study of 4 UK homes with temperature, light, motion, humidity, sound and CO2 sensors, showing how valuable yet expensive sensors are often not chosen in the optimal set.
The reduction of energy use in buildings is a major component of greenhouse gas mitigation policy and requires knowledge of the fabric and the occupant behaviour. Hence there has been a longstanding desire to use automatic means to discover these. Smart meters and the internet-of-things have the potential to do this. This paper describes a study where the ability of inverse modelling to identify building parameters is evaluated for 6 monitored real and 1000 simulated buildings. It was found that low-order models provide good estimates of heat transfer coefficients and internal temperatures if heating, electricity use and CO2 concentration are measured during the winter period. This implies that the method could be used with a small number of cheap sensors and enable the accurate assessment of buildings' thermal properties, and therefore the impact of any suggested retrofit. This has the potential to be transformative for the energy efficiency industry..
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.