From smart homes to smart communities: Advanced data acquisition and analysis for improved sustainability and decision making

The evolving national and international energy policy landscape is resulting in demanding targets for energy demand and greenhouse gas (GHG) reductions along with related socio-economic `triple bottom line' goals. Given this context, and the demographic shift from rural to urban habitation, there is growing concern that current approaches to achieving sustainable built environments and communities are not fit-for-purpose. This paper describes the development of `SmartNet', an adaptable and interoperable `whole-system' framework incorporating a wide-scale data acquisition, analysis and modeling approach that addresses holistic sustainability imperatives. Initially, a candidate architecture for a scaleable domestic condition monitoring and analytic solution is presented, based on key requirements such as cost-effective system delivery in retrofit contexts. Subsequently, a Bayesian probabilistic method is described that facilitates both short term analysis of building energy performance and subsequent system control. Finally, the application of Bayesian inference networks to longer term wider area strategic decision making imperatives is described. The results show that this robust whole-system approach facilitates multi-scale analysis from single building to national levels whilst managing constraints such as cost. Additionally, solutions to problems associated with the management of uncertainty or incomplete datasets, along with issues related to international cooperation and knowledge sharing are presented.